Beruflich Dokumente
Kultur Dokumente
E D I T E D B Y
RI CHARD MARTI N REYCRAF T
THE COTSEN INSTITUTE OF ARCHAEOLOGY AT UCLA
Charles Stanish, Director
Julia L. J. Sanchez, Assistant Director and Director of Publications
Shauna Mecartea, Publications Assistant
This book is set in 10-point Janson Text, with titles in 33-point Meridien.
Edited by Abby Sider
Designed by William Morosi
Library of Congress Cataloging-in-Publication Data
Us and them : archaeology and ethnicity in the Andes / edited by Richard Martin Reycraft.
p. cm.
This volume is a compilation of papers presented in the symposium Us and Them: Archaeology and Ethnicity
in the Andes, held at the 64th Annual Meeting of the Society for American Archaeology in Chicago in April
1999--Indrod.
Includes bibliographical references and index.
ISBN 1-931745-17-X (alk. paper)
1. Indians of South America--Andes Region--Antiquities--Congresses. 2. Indians of South America--Andes
Region--Ethnic identity--Congresses. 3. Indians of South America x Anthropometry--Andes Region--Congresses .
4. Desert archaeology--Andes Region--Congresses. 5. Ethnoarchaeology--Andes Region--Congresses. 6. Human
remains (Archaeology)--Andes Region--Congresses. 7. Grave goods--Andes Region--Congresses. 8. Andes
Region--Antiquities--Congresses. I. Reycraft, Richard Martin. II. Society for American Archaeology. Meeting
(64th : 1999 : Chicago, Ill.)
F2230.1.E84U7 2004
980.00498--dc22
2004025648
Copyright 2005 Regents of the University of California
All rights reserved. Printed in the USA.
DEDICATION
This volume is dedicated to my wife, Soledad, and my parents, Elizabeth and Donald,
whose unending love and support have always inspired and sustained me.
Dedication iii
1. Introduction
Us and Them: Archaeology and Ethnicity in the Andes 1
Richard Martin Reycraft
2. Ethnogenesis at Galindo, Peru 12
Garth Bawden
3. Of Pots and People: Ceramic Style and Social Identity in the Tiwanaku State 34
John Wayne Janusek
4. Style Change and Ethnogenesis among the Chiribaya of Far South Coastal Peru 54
Richard Martin Reycraft
5. Who Were the Priests, the Warriors, and the Prisoners?
A Peculiar Problem of Identity in Moche Culture and Iconography, North Coast of Peru 73
Steve Bourget
6. Household Approaches to Ethnicity on the South Coast of Peru:
The Domestic Architecture of Early Nasca Society 86
Kevin J. Vaughn
7. Cultural Communication of Ethnicity through Clothing:
The Qocha-Lake Symbol in Contemporary Textiles from Ausangate, Peru 104
Andrea M. Heckman
8. North Coast Style after Moche: Clothing and Identity at El Brujo, Chicama Valley, Peru 115
Amy Oakland Rodman and Gioconda Arabel Fernandez Lopez
9. Ethnicity, Kinship, and Ancient DNA 134
Sloan R. Williams
10. A Bioarchaeological Approach to Tiwanaku Group Dynamics 153
Deborah E. Blom
11. A Bioarchaeological Assessment of Prehistoric Ethnicity
among Early Late Intermediate Period Populations of the Azapa Valley, Chile 183
Richard C. Sutter
12. Pescadores and Labradores among the Seoro of Chiribaya in Southern Peru 206
M. C. Lozada and Jane E. Buikstra
13. Migration, Colonies, and Ethnicity in the South-Central Andes 226
Charles Stanish
14. Discussion: Ethnogenesis and Ethnicity in the Andes 233
Jane E. Buikstra
Index 239
Contents
1
T
his volume is a compilation of papers presented in
the symposium Us and Them: Archaeology and Ethnic-
ity in the Andes, held at the 64th Annual Meeting of the
Society for American Archaeology in Chicago in April
1999. All of the original symposium participants have
contributed to this volume. One participant, Andrea
Heckman, was recruited for the Chicago symposium,
but active eldwork prevented her attendance; she has
nevertheless provided a volume chapter.
The symposium assembled a corpus of schol-
ars whose work collectively represents a signicant
advancement in the study of prehistoric ethnicity in the
Andean region. The research presented in the sympo-
sium, and subsequently in this volume, was assembled in
order to represent a diverse collection of theoretical and
methodological approaches. The subsequent chapters
will thus impart recent discoveries in several subelds
of prehistoric Andean anthropology, including spatial
archaeology, textile and ceramic analysis, and, perhaps
most notably, biological anthropology. Many of the
authors in this volume apply novel research -techniques,
while others wield more established approaches in
original ways. Hopefully, their results will stimulate
others to pursue future innovative work in the prehis-
toric study of ethnic identication.
Compared with many other parts of the globe,
the Andean region provides some signicant advan-
tages for the researcher of prehistoric ethnic groups.
Perhaps rst and foremost of these advantages are its
geography and environment. The spectacular Andes
Mountains provide a compressed vertical geography
that juxtaposes fertile coastal and mountain valleys
between inhospitable deserts and mountains. Through-
out Andean prehistory, these valleys were rich breeding
grounds for a mosaic of distinct ethnic groups, and this
diversity provides good comparative data for ethnic
study. Many of these valleys, particularly the moun-
tain variety, are still remote and secluded, which has
provided some insulation against modern development,
allowing traditional cultural practices and dress to per-
sist. While less protected from modern development,
the coastal valleys have extremely arid climates, which
CHAP T E R 1
Introduction
US AND T HE M:
ARCHAE OL OGY AND
E T HNI CI T Y I N T HE ANDE S
GARTH BAWDEN
B A W D E N 13
societies exist in a state of constant change in which
circumstance and historic process conspire to modify
the interests and statuses of individuals and generate
new adaptations and groupings.
In my opinion this diversity of approach illustrates
a most important characteristic of ethnicityits ex-
ibilityfor it can include aspects of all of these forms
of human grouping but is not conned to any one of
them. Thus my usage of ethnicity encompasses only
two basic requirements: rst, recognition of common
membership in a social group in which people con-
sciously share the same values, interests, and goals;
second, the capacity of the group to socially reproduce
itself. Within this self-referential context, ethnicity
not only transcends any single determinant or set of
determinants, it is also historically constructed and
thus changes through time according to situational
circumstance and experiential history.
On the surface this self-ascriptive character of eth-
nicity, its particularity, and its capacity to change with
new circumstances would appear to make its archaeo-
logical investigation difcult if not fruitless. After all,
given the lack of correlation of ethnicity with decorative
or architectural style, or burial custom, there is little
justication to ascribe ethnic signicance to any indi-
vidual or group of material signiers purely on the basis
of their formal characteristicsthe basis of archaeo-
logical analysis. The way past this dilemma is through
the study of social context. Ethnicity, in common with
other social phenomena, does not exist in a vacuum.
Rather it is a manifestation of its location in time and
space. Thus if the context of social production can be
identied, it follows that the related material artifacts
can also be placed within their proper domains of mean-
ing. Such a study requires the reconstruction of the
historic trajectory that created the setting for ethnic
formation as well as the specic social arena within
which it emerged. While by denition this enquiry
must focus on the particular time-space experience of
discrete societies and thus describe a unique historic
circumstance, certain broader developmental forces
offer a productive setting for its realization. Foremost
among these are the scales of integration, measured
by both structural levels and degrees of complexity,
within which ethnic formation occurs.
Scales of Social Integration
At the broadest scale are the effectively unconscious
deep structures that characterize the longue dure of
the worlds great human traditions (Western, Native
American, East Asian, etc.). Although this elemental
structural unity provides the most general determinants
of social afnity, it transcends the assertive local and
regional cultural boundaries within which ethnicity
has meaning. At the active level of human social con-
struction with which I am concerned in this study lies a
complex and reexive interaction between distinct but
overlapping cognitive structures, one relatively innate
and persistent, the other assertive, contingent, yet rela-
tively transitory. At the deepest level of social awareness
all individuals share with their cultural group strong
yet vaguely defined conceptions of self and group,
shaped by their common experience in time and space.
This level of structural cognition that approximates to
Giddenss (1990 [1979]) well-known practical con-
sciousness and Bordieus (1977) habitus constitutes
a given mental language that allows all members of a
society to communicate on both intellectual and emo-
tional levels while providing the broad limits within
which they dynamically interact in the arenas of daily
life. This conceptual language represents the unique
pool of knowledge and belief accumulated by a group
through time. Moreover, it is this shared cognitive stra-
tum that provides the foundation upon which societies
build their specic organizational strategies, whether
characterized by principles of kinship or class in evolu-
tionary parlance.
However, it is at the level of daily interaction that
individuals actively assess their social situation and ques-
tion its utility for their needs. It is also at this level that
the forces that drive ethnogenesis are located. Through
political, social, economic, and ideological contestation
people confront their ideas and policies and strive to
negotiate new outcomes that further their own discrete
goals. Yet such arguments can only be formulated within
the conceptual and experiential milieu provided by the
collective belief structure discussed in the previous sec-
tion. In this regard it is important to note that repeated
negotiation of principles drawn from this shared structure
will inevitably lead to its reexive modication, although
at a much more gradual rate than that associated with
active social decision making.
While it is true that those corporate alliances within
society that control the institutions and economy of
authority are at a distinct advantage in this negotiation
process, subordinate groups are by no means without
recourse. Although frequently separated from the
institutional arena of social negotiation, such politically
secondary groups have access to informal settings of
14 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
community lifehousehold, unofcial or group religion,
marketto afrm their interests and needs. In the nal
analysis dominant and subordinate interests exist in a
state of complementary tension in stable societies. By
contrast, in extreme circumstances of internal stress, the
former exerts its power to protect and, where necessary,
to impose the prevailing social order, while the latter
actively resists. If resolved, group opposition of this form
merely represents a dynamic for social change and need
not lead to the creation of new social identities. How-
ever, an extended period of unresolved internal conict
may create a profound degree of social disruption that
becomes the catalyst for such transformation.
I turn now to the role of social complexity in gen-
erating ethnic awareness. There is clear divergence of
opinion regarding the level of internal social complex-
ity that must be present for ethnicity to emerge. It has
been proposed that self-ascribed identity groupings
only arose in the context of evolving social complexity
and the pressure that this exerts (Bentley 1987). Indeed,
some theorists take this notion so far as to see ethnicity
as a by-product of the modern industrial state (Gell-
ner 1983). From this viewpoint ethnic construction
is understood as a shift in the conception of personal
identity in response to the emergence of new institu-
tions of political, social, and economic domination in
the modern age.
I am very uneasy with the inference that people are
able to purposefully identify and articulate their par-
ticular group afnities only within a context of extreme
social complexity. After all, this leads to the conclusion
that most members of the worlds human population
through history were oblivious to the social dynamics
that impacted their lives, a stance that few scholars
would seriously hold. Consequently, while I accept
that the organizational setting of the parent society
inuences the strategies that are available for ethnic
formation, I believe that traditional cultural evolution-
ary categories like state and prestate or industrial
versus pre-industrial are of minor relevance in under-
standing its motivation. Rather than focusing on overly
general and, to my mind, questionable determinants
to explain ethnic identity, I think it far more meaning-
ful to identify the precise set of forces that impacts
social life in each particular situation. These factors
may include the rise of urbanism, internal social con-
ict, changing systems of production and commodity
exchange, emergence of intersocietal warfare, political
reconstitution, and major climatic disruption. In vari-
ous congurations such forces can initiate the profound
degree of social disturbance that accompanies assertive
realignment of social afnity.
Thus, the crucial element in ethnic construction is
not whether society is grounded in principles of kinship
or class, or within a chiey or state structure. Rather,
it is the effect of changing situations of social life on
individual and group stability. It follows that I do not
restrict ethnic identity to the recent period. Instead, I
assume that it has existed since at least the earliest civi-
lizations and has played a role in dening intergroup
relationships in all such subsequent instances.
In this study, I view the effects on social identity of
major disruption in a society that most scholars would
regard as prestate in universal evolutionary parlance,
where principles of kinship afnity and links to ancestral
land, underpinned by rituals of community and mythic
origin, composed the chief strategies of social integra-
tion. Specically, I examine how an abrupt move from
traditional rural existence to pre-industrial urban life
was the crucial factor that generated ethnic construc-
tion within this persisting organizational setting. I will
show how this process of change with its accompanying
instability caused people to reassess their social circum-
stances. By deliberately and autonomously manipulating
prevailing structural mechanisms, they realigned their
social afnities and created new and powerful founda-
tions of personal and group identity.
Ideology and Social Construction
There is one specic mechanism of ethnic formation
that is of special importance to this study. As I noted in
the previous discussion, all social negotiation occurs in
active daily discourse where human agents reexively
utilize their common cultural consciousness to assert
individual interests. During periods of stability mem-
bers of different social networks and values live in a state
of complementary tension. Their respective principles
are contested within the public sphere as the agendas
of political, corporate, or religious movements and do
not signicantly affect the wider communal identity,
which their adherents all share. However, in extended
periods of instability and discord these beliefs become
the embedded ideological weapons of group conict.
In internal social conict the distinct values and
principles of each contesting group become the asser-
tive ideologies around which its members consolidate.
I here dene ideology as the specialized formulation
of social statement that promotes the aspirations of its
advocates in the wider social arena. As such, ideology
B A W D E N 15
is the possession of all interest groups within a society,
whether dened by corporate, status, or class member-
ship. However, dominant ideology (Abercrombie, Hill,
and Turner 1980)the condensed discourse of ruling
authority and established social orderhas access to the
institutions of political power. Through such participa-
tory mechanisms as pervasive symbolic communication
and ceremonial ritual, the adherents of dominant ideol-
ogy seek to co-opt all members of the community into
accepting the benecial social order that it proclaims.
Indeed, during periods of widespread advantage it may
well be accepted as the unitary belief system of a suc-
cessful way of life and thereby be social reality for all.
In fact, history makes clear that this situation does not
usually persist unchallenged for long. In stable condi-
tions, negotiation and adjustment are contained within
the conceptual boundaries set by dominant ideological
dogma. However, in times of deprivation the perceived
contradiction between political discourse and individual
lived reality can lead to fundamental challenge to the
ruling order by the subordinated groups who are most
vulnerable. If their lived circumstances remain unchanged
they will inevitably confront the existing order with their
own moral and practical agendas. The ensuing contest
may lead to irreparable division between the conicting
social segments, which differentially manipulate their
previously shared cultural heritage to generate new, dis-
tinct, and ultimately irreconcilable ideological constructs.
Over time these originally political concepts consolidate
into exclusive conceptual bases of group identitythe
process of ethnogenesis.
The Symbolism of Social Action
While symbols have been studied from various view-
points (for examples see Firth 1973; Leach 1976; Mach
1993; Ricoeur 1974), most scholars would agree that
they include a wide range of conceptual, behavioral,
and material forms, whose meanings are rooted in the
cognitive world models of their related cultures. Given
the archaeological nature of my topic, I shall conne my
discussion of symbolism to its material aspect. Symbols
codify peoples experience; assert their particular values,
ideologies, and stereotypes; and direct their relations
with others. Moreover they carry an emotional force
that stimulates involvement and mobilizes people to act
on behalf of the communicated message (Turner 1967:
30). These qualities make symbols potent players in the
afrmation of power at all levels of the social ladder and
in the negotiation and construction of group identity.
It is, then, not surprising that they are essential forces
in the central arenas of social control, where they assert
the prestige of the dominant ideology.
However, control of symbolic discourse is not con-
ned to the dominant political sector. While common
cognitive origin ensures that symbols are understood
throughout their community, their potential for exible
interpretation allows subordinate groups to manipulate
symbolic communication in their own interests. This
property becomes important in societies caught in a
state of deep political and social turmoil. As contest-
ing subgroups strive to assert their precedence, and
new allegiances coalesce in response to social stress,
symbolism becomes a powerful vehicle for reinforcing
common awareness among their members and asserting
their specic character and interests in the face of wider
social pressures.
I have to this point emphasized the culture-specic
nature of symbols. However, cultures do not develop
in isolation, but in varying degrees are inuenced by
their neighbors, incorporating foreign elements into
their symbolic systems. Often, intercultural afnity
fosters absorption of such elements so that they are
no longer distinguishable as foreign. That said, it is
not unusual for elite groups to intentionally embrace
foreign symbols in order to enhance their authority and
to elevate them above other sectors of the population.
These adoptions may carry the prestige of admired
exotic images or, more signicantly, of religious or
political meaning that transcends local belief. By their
very nature such symbols resist incorporation into the
broader symbolic structure of society. This tends to
introduce intrasocietal division, which, at a time of
broad disruption, heightens the potential for internal
conict and rejection of the foreign symbols together
with their discredited proponents.
Places carry powerful symbolic meaning, for places
and the people who use them are part of the complex
of physical, behavioral, and conceptual networks that
together constitute the community. The varied expe-
riences that individuals draw from their particular
historic and social locations converge in the places
where social action occurs. There they are actualized
to effect stability or change. The urban built environ-
ment, the setting for this study, comprises various
architectural locations, each shaped so as to promote
the activities of its principal patrons and to symbolically
communicate their social values. Elite architecture
asserts symbolic force on behalf of dominant interests
and includes forms associated with administrative,
16 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
economic, and religious institutions together with the
residences and burial places of the rulers. By contrast
architecture that falls largely under the control of
commoners is almost entirely conned to their homes
and, to a lesser degree, their burials. At the broadest
level residential space actualizes socially appropriate
domestic modes of behavior (Bordieu 1977; Rapoport
1982, 1990:12ff.). However, within these limits house-
hold members, through daily practice, may reexively
manipulate ideas embedded in residential structure to
effect change. By adjusting the symbolic content of
their houses, people manipulate basic residential prin-
ciples to strengthen their collective identity and assert
social solidaritythe essence of ethnicity.
The Archaeological Search for Ethnicity
Where then does this discussion leave us in the search
for ethnicity in archaeology? I believe that we can
circumvent the obvious problems associated with try-
ing to correlate ethnic identity with material style and
form by exploring the dynamics through which it comes
into existence. Examination of the material remains of
human groups in their spatial and temporal contexts can
offer us insight into the social contexts that encourage
the construction of collective identity and the strategies
employed to consolidate it. From this starting point the
physical (and archaeological) symbols of ethnogenesis
nd their logical place as players in the mechanisms of
social production. This archeological strategy has three
important requirements:
1. It must identify the full archaeological context
of change within which ethnic formation occurred in a
society. This requires knowledge of the particular his-
tory of the subject society, including both its internal
development and external connections. It also requires
knowledge of the circumstances that may have stimu-
lated a section of society to reassess its relationship with
the wider community at the time of ethnic formation.
2. It is necessary to determine that social change
visible in the archaeological record was indeed caused
by ethnogenesis. The archaeological record includes
two components that are frequently confused by the
researcher. These include the well-dened symbol-
ism of dominant political ideology and the wider set
of material expressions of communal identity. While
these usually share a common cultural heritage, and
usually overlap in their symbolic content and perceived
meaning within the wider community, they represent
ultimately distinct forms of social integration.
3. In order to be sure that the archaeological changes
do in fact represent a bounded social group, the wider
extracultural archaeological record must be known.
For the remainder of this chapter I apply this
approach to a specic case studythe Late Moche (ca.
AD 600750) town of Galindo in the Moche Valley
of northern Peru. Here rulers and commoners alike
manipulated communal identity, together with its
related visual symbolism and places of social interac-
tion, to consolidate their threatened social positions at
a time of disruption and transformation. In a context of
abrupt urban formation, economic deprivation, social
tension, and political change, the ultimately conict-
ing interests of these two different segments of the
population caused each to seek new principles of social
cohesion within the wider community, a process that
eroded the existing structure of Moche group identity.
ETHNOGENESIS IN THE LATE MOCHE PERIOD
Foundations of Traditional Moche Social Cohesion
Archaeologically, the structure of traditional Moche
social cohesion is most apparent in its great centers
of political control. The tall platform mounds that
still dominate the coastal Peruvian valleys attest to the
ability of rulers to both control the labor of local com-
munities and afrm the dominant ideology of power
through awesome architectural symbols (Moore 1996;
Moseley 1975). Leaders used these articial re-creations
of the rain-giving sacred mountains to fashion an ideo-
logical discourse, by ofciating at familiar mythic rituals
that identied the rulers with these sacred places and
with their embodied meaning (Alva and Donnan 1993;
Bawden 1996; Topic and Topic 1997). Those in power
thereby assumed the responsibility to mediate with the
supernatural to ensure communal health. At the same
time they acquired potent spiritual force to support and
enhance their political status.
In addition to being expressed in architectural forms,
Moche traditional dominant ideology was proclaimed
through portable symbolism. Fine painted ceramic,
metal, and textile objects all carried depictions of the
rituals of power. Themes were usually drawn from
historic local antecedents (Cordy-Collins 1990) but
occasionally they incorporated potent foreign images
into the Moche symbolic system (Menzel 1977:6263).
Striking portrait vessels proclaimed the power of indi-
vidual members of the ruling elite. While the most
elaborate items were reserved for contexts of central
authority, simpler examples of this symbolism circulated
B A W D E N 17
in all arenas of daily life, including the residential. Thus
these symbols endowed the Moche ideological message
with the powerful legitimacy of familiar traditional cul-
tural meaning shared by all strata of society.
Burial practice is especially susceptible to manipu-
lation by intracommunal groups and is an important
vehicle for ideological discourse, as we shall see in this
study. Practices that surround death are in a real sense
conditioned by the social and political needs of the
living, who may manipulate funerary ritual to embel-
lish or actually to construct their social positions and
relationships. The Moche used burial practice to sanc-
tion the dominant social order, to maintain communal
traditions and origins, and to reinforce group integrity
by reference to ancestral lands and the sacred places
associated with them. The traditional Moche mortuary
style was a brick-lined chamber tomb capped by wood
and cane roong, often located in a formal cemetery
(see Donnan 1995; Donnan and Mackey 1978 for
summaries of Moche funerary practices). However,
even when located in a residential area, burials were
placed outside houses (Chapdelaine 1997; Donnan and
Mackey 1978; Topic 1982). People were usually laid on
their backs with small pieces of copper placed at feet,
hands, and mouth, a practice that links Moche ritual to
earlier practice (Fogel 1993:281). Burial offerings often
bore the distinctive symbolism of dominant ideology.
Exceptionally, the tombs of rulers were incorporated
into special burial platforms accompanied by huge
quantities of ne objects. Important aspects of these
traditional practices changed dramatically in Late
Moche times at Galindo.
The character and pervasive social context of tra-
ditional Moche symbolism conformed to a situation
where a related belief system enjoys broad community
support. To a large degree the success of this inclusive
ideology appears to have stemmed from its grounding
in long-term cultural belief in a rural society organized
by principles of communal solidarity. This structural
conception, actualized through ritual and symbolism,
linked people to the essence of their social beings.
Here dominant ideology represented the social reality
of the entire population during a time of political and
economic orescence.
What can we say about Moche ethnic identity in this
earlier period? It is true that a ruling group produced
the ideological symbolism that archaeology terms
Moche culture. But there is no indication that rulers
regarded themselves as ethnically separate from the
wider population with whom they shared the cultural
heritage that inspired this imagery. Indeed, the acces-
sibility of all members of society to the symbols of north
coast cultural heritage during this period archaeologi-
cally supports the presence of an undifferentiated social
awareness in which kinship relations provided the
milieu for unitary social identication.
Crisis on the North Coast
Late in the sixth century AD, north coast Moche poli-
ties experienced profound disruptions that have variously
been ascribed to climatic disaster, external invasion, and
internal tension (see Bawden 1996 for summary). The
impact on the Moche Valley was particularly severe, with
reduction in territorial hegemony (gure 1), economic
decline, and major dune inundation of prime agricultural
land. In response, the focus of settlement and subsistence
shifted inland to the still-irrigable valley neck where the
new town of Galindo was established, replacing the
previous rural pattern of settlement (gure 2; Billman
1996:292). Inevitably, this change threatened the integ-
rity of all social groups, discrediting the long-successful
political order of the rulers while separating commoners
from their ancestral locations, placing them in an alien
urban setting, and disrupting their existing kin-based
community organization. With traditional strategies of
social integration rejected, both rulers and ruled strove
to re-create social stability in the new setting. Ironically,
while both groups manipulated ideological structure in
their endeavors, they did this in very different ways that
led to social division, ethnic formation, and ultimate col-
lapse of the Late Moche system.
Two vital practical requirements for successful domi-
nant ideology are the perception by the populace of its
benets and, in times of stress, the rulers ability to suc-
cessfully explain and contain threats posed by historic
circumstance. There can be little doubt that the long-
lasting earlier Moche system met these needs. However,
it is equally clear that in the conceptual domain, the sev-
enth-century failure caused rejection of the traditional
dominant political ideology together with its ideologi-
cal base and the rituals and symbols that were the active
agents of its authority. Rulers reafrmed their shaken
position by introducing a coercive system of control at
Galindo and adopting new pan-Andean ideological con-
cepts in an attempt to free themselves from identication
with the failures of their predecessors. Unavoidably this
foreign ideology was oriented to the restrictive use of the
elite and excluded the wider population from participa-
tion in the new social order and its rituals.
18 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
0 75
Leche - Motupe
Lambayeque
Zaa
Jequetepeque
Pampa Grande
Chicama
Moche
Galindo
Moche
Vir
Chao
Santa
Nepea
Northern Coast of Peru
Kilometers
Western edge
of mountains
Archaeological
sites
Southern limit of Moche
political control: Moche IV
Southern limit of Moche
political control: Moche V
Figure 1. Map of the north coast of Peru showing the reduction in the area under Moche inuence at the end of the Middle Moche (Moche IV) phase.
B A W D E N 19
Commoners confronted this structural crisis by
manipulating residential ideology to reconstruct a secure
social reality, in so doing largely redening their group
identity. To this end they drew support from their persist-
ing structural beliefsthe principles that had been the
foundations of the rejected traditional Moche ideological
system, which constituted the core of their social expe-
rience. The distinctive ideational systems of ruler and
ruledone universal and looking to maintain political
authority, the other internally focused and promoting basic
group identityrepresented incompatible constructions
of power and ultimately exacerbated the social stress that
characterizes the Late Moche period in the south.
DIVERGING SOCIAL IDENTITIES AND THEIR
SYMBOLS IN THE MOCHE VALLEY
The Galindo Urban Transformation
Replacement of the previous rural settlement pattern
by the town of Galindo (Billman 1996:292) drastically
altered the social environment of the affected popula-
tion and the relationships that structured their lives.
Transition to urbanism demands major modication of
pre-existing community structure, involving new man-
agement institutions and strategies of social organization
(Gailey 1987:3638). Formal rules of social membership,
public order, economic status, and political authority,
together with their accompanying institutions, replace
implicit conventions of local kinship as the predominant
mechanisms of social organization. At Galindo, these
changes emerged in response to widespread political
and economic disruption, marking a major degree of dis-
continuity in Moche customs of social integration that is
archaeologically manifested in the physical remains of the
town. Just as centuries earlier their ancestors had adopted
features of the powerful Chavn religious complex to
confront decline (Burger 1992:183ff.), Galindo rulers
looked beyond their borders for new ideological foun-
dations of authority. They found them in the expansive
highland Wari state, from which they derived symbolic
0 75
Kilometers
Major archaeological sites
Galindo
C.Galindo
R. Moche
Moro
Canal
C.Blanco
Huacas del Sol
and de la Lunda,
Moche
PERU
Chan Chan
Paci f i c Ocean
Figure 2. Map of Moche Valley with sites mentioned in the text.
20 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
elements and possibly a more coercive form of govern-
ment. The town is a massive symbol of the changed social
circumstances (gures 3 and 4), with strict differentiation
of population groups and urban functions apparent in
residential, storage, and governmental architecture.
Residential occupation was segregated according
to size, content, and distance from water source, with
low-status hillside housing being separated from the
rest of the settlement by a massive wall (W1, gure
3) while the small elite residential area, adjacent to
the sites most elaborate public architecture, is also
shielded by tall stone walls (Walls W2 and W3, gure
3). This remarkable emphasis on population segrega-
tion suggests an unprecedented degree of social tension.
Similarly, corporate storage terraces with restricted
access (gures 3 and 4) reect the resolve of rulers
Figure 3. Plan of Galindo.
B A W D E N 21
to control the reduced food resources still available
to them. Indeed, Galindos very location at the valley
neck underscores these pressures. Here its govern-
ment could best utilize the surviving agricultural lands
north of the river and regulate the labor of agricultural
workers relocated to the town. The architecture of
power displays the same qualities of segregation and
control. Three walled complexescercadurasstand in
the center of the town (designated C in gure 3; also
see gure 4). Although differing in detail from Wari
counterparts, they reect the same emphasis on physical
enclosure, functional separation, and exclusivity (Isbell
and McEwan 1991), thereby contrasting with the more
accessible administrative architecture of earlier times,
with its ideological foundations grounded in the com-
mon conceptual history of the entire society.
Figure 4. Galindo from the air. Note the elite burial mound palace enclosure in bottom right corner of photo, small plat form
of traditional form directly above it, and large cercadura enclosures in center and left-center of the photograph. The image (No.
334935) is reproduced by courtesy of the Department of Library Services, American Museum of Natural History.
22 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
The same qualities pervade the most exclusive
form of Galindo elite architecture and contrast with
its antecedents. Sometimes characterized as ances-
tral to the Chim ciudadelas at Chanchan (Complex
P, gure 3; also see gure 4, Conrad 1974:226), this
unique complex comprises a large burial platform with
central chamber, which does not in itself signicantly
differ from the well-known Sipn Royal Tombs of an
earlier time. However, its association with residential
occupation in a walled compound evokes the Wari-
like compartmentalization noted above. Absent from
Galindo as dominant symbols of power are the majes-
tic platform mounds that for centuries had dominated
Moche life. In fact the one edice of this type is little
more than a modication of the ground surface located
away from the central cercaduras and overshadowed by
the new forms (Structure H, gure 3; also see gure 4).
I suggest that it was built primarily to denote a measure
of continuity with the past, but that real power was now
located in the new cercaduras.
Symbols of Ideological Change
The new ideological orientation was also proclaimed
through changes in the portable symbols of author-
ity at Galindo and was primarily visible in decorated
ceramic vessels. Ceramic change took two very different
directions. First, ceramics of traditional form and color
lose the narrative imagery that had depicted the rituals
within which earlier Moche leaders played their power-
enhancing central roles as arbiters with the supernatural
(gure 6). Most drastic was total abandonment of the
portrait vessel (gure 5), which had emphasized the
important political roles of individual leaders (Donnan
2001). With the rejection of this rich tradition of nar-
rative symbolism, Late Moche elite ceramic decoration
at Galindo came to rely largely on abstract geometrical
imagery (gures 7 and 8). Such major modication can-
not simply be ascribed to stylistic development. Rather,
it reects rejection and abandonment of a discredited
ideology and the elimination of its material symbols of
communication.
The second form of ceramic change involved large-
scale innovation. I have described additions to the Late
Moche inventory in full detail elsewhere (Bawden
1994, 2001). Innovations comprise a number of dark
brown and blackware forms, mostly new to the Moche
tradition, produced by oxygen-reduced ring. Most are
embellished with impressed registers of a very restricted
group of abstract geometric designs drawn from earlier
north coast art (gures 913), or are undecorated and
polished. These new forms together with their emphasis
on reduced ring are generally similar to contempo-
rary Wari-inuenced pottery of the central Peruvian
coast (Menzel 1977), while incorporating elements of
regional continuity through their traditional decorative
embellishment (for example see Bankes 1971: gures
7, 8a, 35, 37; Donnan 1978: gure 189; Kroeber 1925:
plates 54h, 55e, 58b; Kutscher 1955: plate 76; Strong
and Evans 1952: plates IX, M, XVI, E; Willey 1953:
plate 59, 1971: gure 3-64).
Thus, it appears that Moche Valley leaders trans-
formed the nature of symbolic communication in
order to assert a new exclusive ideology of power. They
eliminated the narrative iconographic scheme that had
Figure 5. Portrait Vessel, Middle Moche period (after Fiedel 1987:
gure 107c.)
B A W D E N 23
played a central role in the visual discourse of earlier
ideology, they retained purely decorative traditional
elements of ceramic art as passive symbols of cultural
continuity, and they selectively adopted elements that
evoked the transcendental prestige of a powerful pan-
Andean religious belief system. This symbolic system
and its associated tenets attempted to visually mediate
transformation with continuity. However, given the
cultural and historic situation of its proponents, these
basically incompatible forces posed a daunting chal-
lenge to the new ideology.
Late Moche leaders sought to free themselves from
the localizing strictures of the discredited political past
by separating their authority from wider community
sanction and linking it with a foreign ideational sys-
tem. It is possible that the residual links with the past
expressed in the limited use of older iconographic motifs
indicate attempts to pacify commoners by reference to
a shared cultural past. However, such tokens could not
mask the reality that large segments of the population
were effectively removed from any signicant partici-
pation in the institutions and rituals that dened their
social existence, a drastic change from earlier Moche
periods. The alienating effects of this disruption were
naturally augmented by the relocation of people from
their ancestral lands to urban Galindo, with its enforced
social stratication, economic deprivation, and threats
to traditional communal life. Commoners were faced
with the unprecedented challenge of maintaining their
social beings in an adverse setting. Moreover, they found
themselves in a situation of great political asymmetry,
with all effective power in the hands of their rulers.
The associated tensions collectively created a profound
degree of structural contradiction within the newly
urban society that generated a process of differentiation
between the social identities of rulers and ruled.
THE MECHANISMS OF ETHNIC FORMATION
The Archaeological Setting
At Galindo, people manipulated the limited social arena
at their disposalthe residential householdto con-
struct new forms of ethnic identity on the threatened
residue of the old. The residential house contained
three internal walled spaces: a food preparation area
(cocina, gure 14), a living area (sala, gure 14), and a
Figure 6. Narrative ne line ceramic painting, Middle Moche period.
24 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Figure 7. Traditional ceramic orero of nonrepresentational design. From high-status residential context.
Figure 8. Traditional ceramic bowl of nonrepresentational design. From high-status residential context.
B A W D E N 25
small domestic storage area (deposito, gure 14). The
rectangular living space, characterized by stone-faced
benches lining the walls and enclosing a small patio,
was the center of domestic social activity. However,
this space served another important purpose. Galindo
commoners were buried here within the benches in an
unprecedented form of in-house funerary ritual, which
signals their endeavor to create their own distinc-
tive social practice. Burials were generally modest in
form and content. They usually comprised simple pits
located in the bench cores, but sometimes included a
roughly lined adobe or stone-walled chamber (gures
15, 16), a design that reveals their identity as formal
interments rather than casual disposals. In all cases the
related bench surface contained evidence of general
domestic activities, indicating that burial occurred
during the active life of the house. The deceased were
wrapped in simple textiles and either rested on their
backs, or in contrast with earlier times, lay partially
exed on their sides. Small pieces of copper placed at
head and feet marked a rare feature of continuity with
the north coast funerary tradition (Fogel 1993:281).
Unlike elite funerary practice at Galindo itself, and in
a dramatic departure from earlier Moche practice, no
ceramic items of any kind were incorporated into the
resi dential burials.
Elite burial practice offers a vivid contrast to this resi-
dential pattern. The dead were interred in rectangular
stone-lined chamber tombs accompanied by abundant
ne ceramics and textiles that embodied the foreign-
derived ideological symbolism described above. Burials
were placed in small cemetery clusters, located near
the new architectural centers of authority, the cercaduras.
On the surface this practice appears to follow Moche
cultural convention while the adoption of residential
interment by the Galindo commoners breaks with all
existing regional traditions. Closer scrutiny, however,
suggests that the reverse is true. I have already stressed
that the rulers of Galindo adopted new iconographic
and architectural components in order to break with
and transcend the discredited symbols and practices of
the past. The incorporation of the material signs of this
transformation into the highly charged symbolic setting
of burial was part of this attempt to transform the social
order through dominant ideological modication. In
contrast the residential customs reveal the attempt of a
newly urbanized subject population to reconstruct group
identity from traditional social principles.
Burial as Social Identity
The treatment of the dead carries great potential for
furthering the aspirations of the living through the
manipulation of historically constituted cultural and
religious belief (for various recent archaeological
studies see Hodder 1990; Kristiansen 1991; Parker
Pearson 1993; Thomas 1993; Tilley 1996). Funerary
custom represents a powerful component of social
practice. It directs relationships between members of
a society, living and dead, and it mediates the status
of individuals or groups within a community (Turner
1967; van Gannep 1960 [1909]). Understanding this
potent capacity for social action helps us to explore the
paradoxical juxtaposition of major material change and
conceptual continuity embodied in Galindo residential
burial practices.
We must rst understand the historically consti-
tuted social context within which the Galindo changes
occurred. Earlier north coast social organization, in
accord with broader Andean practice, was grounded in a
pattern of local kinship groups strongly attached to their
ancestral lands through a combination of mythic asso-
ciation and sacred geography. It is probable that prior
to the seventh-century disruptions the Moche Valley
population was loosely integrated within a segmentary
system in which groups retained social autonomy and
communal land privileges and owed labor obligations
to their local and wider political leaders (Moseley 1975;
see Netherly 1984, 1990 for somewhat later Chim
organization). At the core of this system was, and is,
the implicit acceptance that human groups existed at the
conjuncture of their spatial and temporal experience.
Thus the sacred space of communal lands represented
the map on which social life was distributed. On the
temporal plane, life existed as a projection of the past,
continually reinterpreted through myth and its ritual
actualization. In this worldview time, space, and social
reality were inextricably enmeshed, and occupation of
ancestral lands was just one unquestioned denominator
of human social being.
Taking this discussion a step further, we see that
ancestral spirits were active players in the daily affairs
of the living. Today this tenacious principle is regularly
manifested through ritual in which traditional healers
enter sacred time and space to mediate with ancestral
and natural forces to attain the cosmological balance
necessary for social order (Joralemon and Sharon
1993). Moreover, such ritual inevitably involves places
charged with spiritual force, which dene the geo-
graphical experience of the living and connect them,
26 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Figure 9. Innovative reduced ware double-chambered vessel from high-status burial.
Figure 10. Fragment of innovative reduced ware square bowl from high-status residential context.
B A W D E N 27
Figure 11. Innovative reduced ware vessel with elliptical-shaped body. From high-status residential context.
Figure 12. Innovative reduced ware angle-rimmed bowl. From high-status residential
context.
Figure 13. Innovative reduced ware round bowl. From low-status residential context.
28 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Figure 14. Plan of low-status residential structure located at base of Cerro Galindo.
B A W D E N 29
their ancestors, and their descendants with their wider
natural and spiritual world. It is not surprising, then,
that in the pre-Christian epoch the physical remains of
the ancestral dead were awarded special treatment and
were themselves regarded as sacred beings, or huacas.
The disruptions of the Late Moche period threat-
ened these basic structural foundations of daily life.
By congregating numerous distinct communities in
a common location, urbanism, with its new institu-
tional regulations and social compartmentalization,
destroyed the collective boundaries and threatened the
very social identities of many of Galindos inhabitants.
In the new restrictive environment they were alienated
from the sacred places that brought social cohesion
to their daily lives through their innate physical and
spiritual signicance. The narrowly focused political
ideology of their rulers, intended as it was to consoli-
date an exclusive position of authority, could not begin
to replace the lost foundations of group cohesion. This
rupture with their traditional way of life left common-
ers with a profound need to redene and integrate
sacred landscape and temporality in a radically differ-
ent physical context. People responded to this threat
by reconstructing the familiar substance of their social
identity in the new setting through innovative funerary
practice and symbolism.
At Galindo the only location that could be largely
controlled by the non-elite population was the domes-
tic residence. Given their deep connection to the most
tenacious structural manifestations of community life,
residential spaces and their architectural styles always
play fundamental roles in shaping the collective expe-
rience of their builders. At Galindo the low-status
residents transformed this conned domestic social
setting into the reduced counterpart of the wider
sacred landscape from which they had been displaced.
Here they created a system of ritual practice and belief
whose physical expressions ignored reference to domi-
nant discourses of Moche power, past or present. In the
relative autonomy of their homes, they eliminated from
their material culture the narrative symbolism that had
previously carried the tenets of dominant ideological
discourse into all social arenas of the Moche domain.
Indeed the only note of continuity was struck by the
persistence of traditional forms intentionally devoid
of such ideological communication. Even here, in the
domestic arena, new blackware bowls hastened replace-
ment of this tenuous symbolic link to the past. In this
they shared with their rulers rejection of the symbolism
of a failed ideology.
However, unlike their rulers, the commoner popula-
tion of Galindo also virtually banished the symbolism of
the new dominant ideology from their domestic space
and burial ritual. Ceramics embellished with the visual
symbols of the transcendental Late Moche ideology
of power are largely absent from low-status domestic
space. More dramatically, commoner burials diverged
in every way from traditional customin form, loca-
tion, position of the dead, and grave goods. Moreover,
the wide variety of positions displayed by the interred
Figure 15. Adult male burial within room bench of low-status residential structure.
Figure 16. Child burial within room bench of low-status
residential structure.
30 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
dead suggests that the physical setting of the burials
played a more important role in afrming the meaning
of residential funerary practice than did formal organi-
zation or content. This attribute better reects a newly
established practice created at the informal household
level than one dictated by long-term historic custom.
With the pervasive elimination of the material
manifestations of all dominant Moche ideologies from
their domestic lives, lower-status inhabitants of Galindo
signaled their alienation from both past and present
social order. In response people drew on the deep his-
toric foundations of their cultural existence to create a
unique form of social identity. They inserted the bod-
ies of their dead ancestors into the pivotal social space
of the household as huacas and spiritual presences.
Thereby they reconstituted the elemental relation-
ships that had historically formed the core of their social
beings, through a process that involved cultural inven-
tion and reexive interaction between themselves and
the exclusive ideological meaning embedded in their
new household symbolic and ritual practices. Although
associated household funerary activities are not acces-
sible to the archaeological record, we can assume that
they added the social domain within which households
redened their identities in daily action. Here in liminal
time they gathered together the various threads that
provided cohesion and continuity to their social lives. It
is true that the experience underpinning these innova-
tions was embedded in traditional cultural conception.
However, such radical change, which affected an
entire subgroup, signals the presence of fragmentation
within the wider social body that cannot be explained as
temporary adjustment related to the evolutionary emer-
gence of urbanism. Rather it represents the rise of an
entirely new social group, with its own set of values and
practices as the result of a process of ethnogenesis.
CONCLUSION
My interpretation of the Galindo archaeological
recordas the record of ethnogenesis conducted in the
face of severe social transformation at the end of the
Moche periodmakes understandable some otherwise
puzzling contradictions. Ironically, the very different
forms of material symbolism used by rulers and com-
moners can be attributed to a common determination
to maintain identity through change. However, deeper
examination of this shared motivation shows that it took
two very different directions according to the divergent
pressures placed on its proponents by the threatening
events that took place at the end of the seventh century
AD. Political leadership is always dened and asserted
through its use of ideology communicated through
symbolic communication. It follows that at a time of
broad-based breakdown, rulers strove to sustain their
positions as arbiters of power by distancing themselves
from their predecessors failure and imposing a new ide-
ology, bereft of the negative connotations of the past. To
do this they drew on conceptual and symbolic signiers
that transcended local cultural experience to reconstitute
prestige, social distance, and authority on new untar-
nished foundations. Here their concern was maintenance
of their political status, not underlying social identity, and
the means of accomplishing this was effective ideological
change in the context of urban formation.
In contrast, commoners were faced with a very differ-
ent, though more fundamental, challenge. The process
of relocation from their communal lands to Galindo
brought disruption to their earlier, kinship-based way
of life, thus to the traditional foundations of their social
cohesion. Consequently, in the hostile urban setting of
Galindo they were motivated by the need to re-establish
social identity rather than to strive for status or politi-
cal authority like the ruling elite. Unlike their rulers
they eschewed assertive ideology for ethnogenesis as
the means to ensure group cohesion. The paradoxical
blend of material innovation and conceptual continu-
ity apparent in their residential practices denotes this
process of ethnic construction. Within the familiar
and secure setting of the household, commoners used
familiar cultural principles to reconstitute the essential
tenets of historic north coast social consciousness in
the urban context. At the same time they narrowed and
redened the effective scope of these powerful cultural
concepts from their previous role as the core principles
of an entire inclusive society to one in which they served
as anchors of a particular threatened group. The long-
familiar beliefs and practices that had been the basis for
broad social identity and dominant ideology alike were
thereby co-opted by a threatened group to actively cre-
ate a new sense of identity in the restricted household
context. While the rough household tombs can easily
be regarded as benign alternatives to the burials of the
elite, in fact they embodied active separation, resistance,
group autonomy, and excusive social cohesion.
Inevitably as part of this process traditional Moche
symbolic imagery, with its identication with a discarded
past, was rejected by all segments of society as a means of
maintaining group identity, albeit for different reasons.
On one hand, rulers replaced the rejected traditional
B A W D E N 31
forms with the symbols of a more effective ideology,
introducing new material agents of communication for
this purpose. On the other hand, as part of a necessary
process of social renewal, commoners strove to eliminate
all traces of dominant Moche symbolism from the house-
hold setting. Both moves furthered the construction of
competing social identities whose material symbols are
already at Galindo difcult to characterize as Moche.
In the diverse symbolism that weakly evoked a rapidly
receding past and more powerfully reached toward an
uncertain future, we see the irrevocable beginnings of
the ethnic transformation that would ultimately produce
Chim cultural identity. In conclusion, Galindo provides
us with an unusually advantageous case study of ethnic
formation. By recovering the full social context of the
human group that experienced these developments as
the convergence of unique and threatening spatial and
temporal forces, we are able to distinguish the signiers
of ethnic assertion from those of ideological authority
and other ambiguous social referents. By so doing we are
able to view the mechanisms of social production in their
true character as the actions of individuals addressing the
adverse circumstances in which they found themselves at
a pivotal moment in time by creating new foundations
of group cohesion.
REFERENCES
Abercrombie, N., S. Hill, and B. S. Turner
1980 The dominant ideology thesis. London: George Allen
and Unwin.
Alva, W., and C. B. Donnan
1993 Royal tombs of Sipn. Los Angeles: Fowler Museum
of Cultural History, University of California.
Blint, C.
1994 Some ethnospecic features in central and eastern
European archaeology during the early Middle Ages:
The case of Avars and Hungarians. In Archaeological
approaches to cultural identity, edited by S. J. Shennan,
185193. London: Routledge.
Bankes, G. H. A.
1971 Some aspects of the Moche culture. Ph.D. disserta-
tion, Institute of Archaeology, London University.
Bawden, G.
1994 Nuevas formas de cermica Moche V procedentes de
Galindo. In Moche: Propuestas y perspectivas, edited by
S. Uceda and E. Mujica, 207222. Actas del Primer
Coloquio sobre la Cultura Moche (Trujillo, 12 al
16 de abril de 1993). Travaux de lInstitut Franais
dEtudes Andines 79. Lima.
1996 The Moche. Oxford: Blackwell Publishers.
2001 The symbols of late Moche social transformation.
In Moche art and archaeology in ancient Peru, edited by
J. Pillsbury, 285306. New Haven: Yale University
Press.
Bentley, G. C.
1987 Ethnicity and practice. Comparative Studies in Society
and History 29:2455.
Billman, B. R.
1996 The evolution of prehistoric political organizations in
the Moche Valley, Peru. Ph.D. dissertation, Depart-
ment of Anthropology. University of California,
Santa Barbara. University Microlms, Ann Arbor.
Bordieu, P.
1977 Outline of a theory of practice. Cambridge: Cambridge
University Press.
Burger, R.
1992 Chavn and the origins of Andean civilization. London:
Thames and Hudson.
Chapdelaine, C.
1997 Le tissus urbain du site Moche. In lombre du
Cerro Blanco, edited by C. Chapdelaine, 1182. Les
Cahiers danthropologie 1. Montreal: University of
Montreal.
Conkey, M., and C. Hastorf
1990 The uses of style in archaeology. New York: Cambridge
University Press.
Conrad, G. W.
1974 Burial platforms and related structures on the north
coast of Peru: Some social and political implications.
Ph.D. dissertation, Department of Anthropology,
Harvard University.
Cordy-Collins, A.
1990 Archaism or tradition? The decapitation theme in
Cupisnique and Moche iconography. Latin American
Antiquity 3 (3): 206220.
Donnan, C. B.
1978 Moche art of Peru. Los Angeles: Museum of Cultural
History, University of California.
1995 Moche funerary practice. In Tombs for the living:
Andean mortuary practices, edited by T. Dillehay,
111160. Washington, D.C.: Dumbarton Oaks
Research Library and Collection.
2001 Moche ceramic portraits. In Moche art and archaeol-
ogy in ancient Peru, edited by J. Pillsbury, 127140.
Washington, D.C., and New Haven: National Gal-
lery of Art and Yale University Press.
Donnan, C. B., and C. J. Mackey
1978 Ancient burial patterns of the Moche Valley, Peru. Aus-
tin: University of Texas Press.
32 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Fiedel, S. J.
1987 Prehistory of the Americas. Cambridge: Cambridge
University Press.
Firth, R.
1973 Symbols, public and private. London: George Allen
and Unwin.
Fish, S.
1989 Doing what comes naturally. Durham, NC: Duke
University Press.
Fogel, H.
1993 Settlements in time: A study of social and political
development during the Gallinazo occupation of the
north coast of Peru. Ph.D. dissertation, Department
of Anthropology, Yale University. University Micro-
lms, Ann Arbor.
Foucault, M.
1984 Truth and power. In The Foucault reader, edited by
P. Rabinow, 5176. London: Pantheon Books.
Gailey, C. W.
1987 Culture wars: Resistance to state formation. In Power
relations and state formation, edited by C. W. Gailey
and T. C. Patterson, 3556. Salem, Wisconsin: Shef-
eld Publishing Company.
Gellner, E.
1983 Nations and nationalism. Oxford: Basil Blackwell.
Giddens, A.
1990 [1979] Central problems in social theory. Berkeley:
University of California Press.
Harman, W. W.
1988 The postmodern heresy: Consciousness as casual.
In The reenchantment of science: Postmodern proposals,
edited by D. R. Grifn, 123151. Albany: State
University of New York Press.
Hodder, I.
1990 The domestication of Europe. Oxford: Blackwell Pub-
lishers.
Hubert, J.
1989 A proper place for the dead: A critical review of the
reburial issue. In Conict in the archaeology of living
traditions, edited by R. Layton, 129164. London:
Unwin Hyman.
Isbell, W. H., and G. F. McEwan, eds.
1991 Huari administrative structure: Prehistoric monumental
architecture and state government. Washington, D.C.:
Dumbarton Oaks.
Joralemon, D., and D. Sharon
1993 Sorcery and shamanism: Curanderos and clients in north-
ern Peru. Salt Lake City: University of Utah Press.
Kristiansen, K.
1991 Chiefdoms, states, and systems of social evolution.
In Chiefdoms: Power, economy, and ideology, edited by
T. Earle, 1643. Cambridge: Cambridge University
Press.
Kroeber, A. L.
1925 The Uhle pottery collection from Moche. University of
California Publications in American Archaeology
and Ethnology 21 (5): 191234. Berkeley: Univer-
sity of California Press.
Kutscher, G.
1955 Ancient art of the Peruvian north coast. Berlin: Gebr.
Mann.
Leach, E. R.
1976 Culture and communication. Cambridge: Cambridge
University Press.
Mach, Z.
1993 Symbols, conict, and identity. Albany: State University
of New York Press.
Menzel, D.
1977 The archaeology of ancient Peru and the work of Max
Uhle. Berkeley: R. H. Lowie Museum of Anthropol-
ogy, University of California.
Moore, J. D.
1996 Architecture and power in the Ancient Andes: The
archaeology of public buildings. Cambridge: Cambridge
University Press.
Moseley, M. E.
1975 Prehistoric principles of labor organization in
the Moche Valley, Peru. American Antiquity 40:
191196.
Netherly, P.
1984 The management of late Andean irrigation systems
on the north coast of Peru. American Antiquity 49:
227254.
1990 Out of the many, one: The organization of rule in
the north coast polities. In The northern dynasties:
Kingship and statecraft in Chimor, edited by M. E.
Moseley and A. Cordy-Collins, 461487. Washing-
ton, D.C.: Dumbarton Oaks.
Parker Pearson, M.
1993 The powerful dead: Archaeological relationships
between the living and the dead. Cambridge Archaeo-
logical Journal 3 (2): 203229.
Rapoport, A.
1982 The meaning of the built environment: A nonverbal
communication approach. Beverly Hills: Sage Publi-
cations.
1990 Systems of activities and systems of settings. In
Domestic architecture and the use of space: An interdis-
ciplinary cross-cultural study, edited by S. Kent, 920.
Cambridge: Cambridge University Press.
B A W D E N 33
Ricoeur, P.
1974 The conict of interpretations. Evanston, IL: North-
western University Press.
Sperber, D.
1975 Rethinking symbolism. Cambridge: Cambridge Uni-
versity Press.
Strong, W. D., and C. Evans
1952 Cultural stratigraphy in the Vir Valley, northern Peru.
Columbia Studies in Archaeology and Ethnology 4.
New York: Columbia University Press.
Thomas, J.
1993 The hermeneutics of megalithic space. In Interpreta-
tive archaeology, edited by C. Tilley, 7398. Oxford:
Berg Publishers.
Tilley, C.
1996 An ethnography of the Neolithic: Early prehistoric societ-
ies in southern Scandinavia. Cambridge: Cambridge
University Press.
Topic, J. R., and T. L. Topic
1997 La guerra Mochica. Revista Arqueolgica SIAN,
Edicin 4, revista semestral, November: 1012.
Topic, T.
1982 The Early Intermediate period and its legacy. In
Chan Chan: Andean desert city, edited by M. E.
Moseley and K. Day, 255284. Albuquerque: Uni-
versity of New Mexico Press.
1991 The Middle Horizon in northern Peru. In Huari
administrative structure: Prehistoric monumental archi-
tecture and state government, edited by W. H. Isbell
and G. F. McEwan, 233246. Washington, D.C.:
Dumbarton Oaks.
Trigger, B. G.
1989 A history of archaeological thought. Cambridge: Cam-
bridge University Press.
Turner, V.
1967 The forest of symbols. Ithaca, NY: Cornell University
Press.
Ucko, P.
1969 Ethnography and archaeological interpretation of
funerary remains. World Archaeology 1:262280.
van Gannep, A.
1960 [1909] The rites of passage. Chicago: University of
Chicago Press.
Willey, G. R.
1953 Prehistoric settlement patterns in the Vir Valley, Peru.
Bureau of American Ethnology Bulletin 155. Wash-
ington, D.C., Smithsonian Institution.
1971 An introduction to American archaeology. Vol. 2, South
America. Englewood Cliffs, NJ: Prentice Hall Inc.
34
INTRODUCTION
Understanding the emergence and consolidation of
past states requires that we address social identity.
Most archaeologists skirt the problem because of the
difculties involved in relating identity to material cul-
ture, or of observing identity in the dirt. Ceramics, in
many societies some of the most expressive material
remains, are also the most anathematized of artifacts
used to approach past identity, in part a response to
early idealistic attempts to relate ceramic style with
residential and kinship patterns (Hill 1970; Longacre
1964). The abundant critiques are well founded and
we are armed with a wealth of cautionary tales (e.g.,
Dietler and Herbich 1998; Hodder 1982), but we now
face an aversion to addressing material culture as an
expression of social identity at all. The conceptual
bath water needs reconsidering, but we have prema-
turely discarded some of the most important potential
expressions of human activity.
I examine material expressions of group identity
and their role in the prehispanic state of Tiwanaku
(AD 5001150), a polity centered in the southern Lake
Titicaca basin of contemporary Bolivia (Albarracin-
Jordn 1992, 1996; Alconini Mjica 1995; Bermann
1990, 1994; Janusek 1994; Kolata 1991, 1993; Man-
zanilla 1992; Mathews 1992; Rivera Casanovas 1994;
Stanish 1994). I argue that certain classes of ceramic
vessels in the ancient Lake Titicaca basin, through
both their crafting and use, became involved in the
expression and negotiation of status and social iden-
tity. What we require in order to see identity in the
past is a cautious, multidimensional approach to the
relation between human action and material culture. I
compare excavated residential compounds in the urban
centers of Tiwanaku and Lukurmata. I examine patterns
of conformity and differentiation in ceramic style in
relation to other aspects of residential life, including
spatial organization, craft production, diet, and burial
patterns. A conjunction of material patterns, represent-
ing multiple dimensions of practical consciousness and
social activity, point to the central place of identity
in the organization and day-to-day formulation and
reproduction of Tiwanaku society. Ceramic vessels
were signicant vehicles not only for the production,
CHAP T E R 3
Of Pots and People:
CE RAMI C S T Y L E AND S OCI AL
I DE NT I T Y I N T HE T I WANAKU S TAT E
a
d
o
m
e
s
t
i
c
s
t
r
u
c
t
u
r
e
s
(
r
e
d
r
a
w
n
f
r
o
m
C
o
n
r
a
d
1
9
9
3
)
.
R E Y C R A F T 69
F
i
g
u
r
e
1
2
.
T
e
r
m
i
n
a
l
p
h
a
s
e
C
h
i
r
i
b
a
y
a
d
o
m
e
s
t
i
c
s
t
r
u
c
t
u
r
e
.
70 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
I suggest that the Terminal phase Chiribaya people
had begun the process of changing their ethnic identi-
cation. Exterior clothing and ritual ceramic wares, the
items of material culture that have the most emblematic
potential, had borrowed signicant Estuquia stylistic
attributes. Vernacular architecture and tomb form, more
latent indicators of ethnic identication, had not. The
Inka, who added the Osmore region to their empire in
the early fourteenth century, may have fueled this ethnic
genesis. In the lower Osmore, the Inka archaeological
footprint consists of a scant few imperial ceramics and
some indigenous Chiribaya/Chiriquia ceramic wares
with Inka stylistic attributes, such as beveled bowl rims
(Reycraft 1998). No Inka administrative centers, settle-
ments, or tambos
3
have been found in the lower Osmore.
In contrast, an Inka administrative center and tambo have
been found in the upper Osmore area (Burgi 1993), and
Late Horizon Estuquia ceramic wares display so many
Inka attributes that they are classied as Estuquia-Inka
wares. The Inka clearly administered the Osmore from
the upper valley, and for administrative purposes they
may have encouraged the Chiribaya to ethnically afli-
ate with the Estuquia by classifying both groups into
a single administrative category. Alternatively, the Inka
may have recognized the Terminal Chiribaya as a sepa-
rate people who lived in an impoverished, depopulated
region that required little administration. The presence
of a ruling elite in the upper valley, and the prestige rela-
tionships that it engendered, could have also fueled the
Chiribaya adoption of Estuquia ethnic paraphernalia.
Unfortunately, we can never know the end result of this
reidentication process, for it was cut short by the Span-
ish colonization.
The Spanish conquest of the Andes had a profound
effect on the ethnic groups of the region. War, disease,
and excessive tribute demands made by the Spanish
resulted in catastrophic mortality and migration pat-
terns throughout vast areas of the Andes. Dramatic
population shifts disrupted many of the traditional
ayllu reciprocal relationships upon which crafts people
were dependent for procurement of raw materials for
the production of their material culture. Finally, dur-
ing the 1780s the Spanish authorities established dress
laws for indigenous peoples, whereby much traditional
ethnic dress was outlawed and European-style pants,
vest, and a jacket deemed mandatory.
Given the opportunity to continue their trends,
the postdisaster Chiribaya people may have been
assimilated into a broader Estuquia ethnic group.
Alternatively, they may have only borrowed specic ele-
ments of the latter culture and reidentied themselves
as a separate ethnic entity, one whom we would possibly
term Chiriquia. All that we can reasonably assume,
however, is that the postdisaster Chiribaya people did
not view their ethnicity in the same manner as their
Classic phase ancestors.
NOTES
1. A fog-fed vegetation, indigenous to the coastal Andean
region.
2. Chulpas are circular tower tombs that were constructed
for the elite in regions of Aymara cultural inuence.
3. A tambo was a way station along an Inka road. Tambos
served communications and supply purposes.
REFERENCES
Aldenderfer, M. S., and C. Stanish
1993 Domestic architecture, household archaeology,
and the past in the south-central Andes. In Domes-
tic architecture, ethnicity, and complementarity in the
south-central Andes, edited by M. S. Aldenderfer,
112. Iowa City: University of Iowa Press.
Barth, F.
1969a Introduction. In Ethnic groups and boundaries: The social
organization of culture difference, edited by F. Barth,
938. Boston: Little, Brown and Company.
1969b Pathan identity and its maintenance. In Ethnic groups
and boundaries: The social organization of culture dif-
ference, edited by F. Barth, 117134. Boston: Little,
Brown and Company.
Bawden, G.
1989 Pre-Incaic cultural ecology of the Ilo region. In
Ecology, settlement, and history in the Osmore Drainage,
Peru, edited by D. Rice, C. Stanish, and P. R. Scarr,
183206. BAR International Series 545 (i). Oxford:
British Archaeological Reports.
1993 An archaeological study of social structure and eth-
nic replacement in residential architecture of the
Tumilaca Valley. In Domestic architecture, ethnicity,
and complementarity in the south-central Andes, edited
by M. S. Aldenderfer, 4254. Iowa City: University
of Iowa Press.
Boyd, R., and P. J. Richerson
1987 The evolution of ethnic markers. Cultural Anthropol-
ogy 2:6579.
Boytner, R.
1992 Ilo-Tumilaca/Cabuza, Chiribaya, and their relation-
ships with the Tiwanaku Culture as can be seen in the
R E Y C R A F T 71
textiles from the coastal Osmore Valley, southern Peru.
Masters thesis, Institute of Archaeology, University of
California, Los Angeles.
Buikstra, J.
1995 Tombs for the livingorfor the dead: The
Osmore ancestors. In Tombs for the living: Andean
mortuary practices, edited by T. Dillehay, 229280.
Washington, D.C.: Dumbarton Oaks Research
Library and Collection.
Burgi, P. T.
1993 The Inka Empires expansion into the coastal
Sierra region west of Lake Titicaca. Ph.D. disserta-
tion, Department of Anthropology, University of
- Chicago.
Childe, V. G.
1929 The Danube in prehistory. Oxford: Clarendon Press.
1933 Races, peoples, and cultures in prehistoric Europe.
History 18:193203.
1935 Changes, methods, and aims in prehistory. Presidential
address in Proceedings of the Prehistoric Society 1:115.
Cieza de Len, P. de
1986 [1553] La crnica del Per. 3 vols. Lima: Pontica Uni-
versidad Catlica del Per y Academia Nacional de
la Historia.
Clark, N. R.
1993 The Estuquia textile tradition: Cultural pattern-
ing in the late prehistoric fabrics of Moquegua, far
southern Peru. Ph.D. dissertation, Department of
Anthropology, Washington University, St. Louis.
Clark, N. R., P. Palacios, and N. Juarez
1993 Proyecto textil III, Chiribaya Baja: Cementerio 1 fardos
y textiles. Internal Report of Proyecto Contisuyu.
Moquegua, Peru: Museo Contisuyu.
Cobo, B.
1956 [1653] Historia del nuevo mundo. Edited by P. F. Mateos.
Biblioteca de Autores Espaoles 9192. Madrid:
Ediciones Atlas.
Cohen, A.
1974 Introduction: The lesson of ethnicity. In Urban
ethnicity, edited by A. Cohen, ix-xxiii. London:
Tavistock Publications.
Conrad, G. W.
1993 Domestic architecture of the Estuquia Phase at
Estuquia and San Antonio. In Domestic architec-
ture, ethnicity, and complementarity in the south-central
Andes, edited by M. S. Aldenderfer, 5565. Iowa
City: University of Iowa Press.
Eidheim, H.
1969 When ethnic identity is social stigma. In Ethnic
groups and boundaries: The social organization of culture
difference, edited by F. Barth, 3957. Boston: Little,
Brown and Company.
Gasparini, G., and L. Margolies
1980 Arquitectura Inka. Bloomington: Indiana University
Press.
Geertz, C.
1963 The integrative revolution: Primordial sentiments
and civil politics in the new states. In Old societies and
new states, edited by C. Geertz, 105157. New York:
The Free Press.
Glazer, N., and D. P. Moynihan
1975 Introduction. In Ethnicity: Theory and experience,
edited by N. Glazer and D. P. Moynihan, 126.
Cambridge, MA: Harvard University Press.
Haaland, G.
1969 Economic determinants in ethnic processes. In Eth-
nic groups and boundaries: The social organization of
culture difference, edited by F. Barth, 5873. Boston:
Little, Brown and Company.
Hodder, I.
1979 Economic and social stress and material culture
-patterning. American Antiquity 44 (3): 446454.
Hyslop, J.
1976 An archaeological investigation of the Lupaqua king-
dom and its origins. Ph.D. dissertation, Department
of Anthropology, Columbia -University.
1989 Inka settlement planning. University of Texas Press.
Issacs, H.
1974 Basic group identity: Idols of the tribe. Ethnicity 1:
1541.
Jessup, D.
1990 Desarollos generales en el intermedio tardio en el Valle
de Ilo, Per. Internal Report of Proyecto Contisuyu.
Moquegua, Peru: Museo Contisuyu.
1991 General trends in the development of the Chiribaya
culture, south coastal Peru. Paper presented at the
1991 meeting of the Society for American Archaeol-
ogy, New Orleans.
Jones, S.
1997 The archaeology of ethnicity: Constructing identities in
the past and present. London and New York: Rout-
ledge Press.
Julien, C.
1983 Hatunqolla: A view of Inka rule from the Lake Titicaca
region. Series Publications in Anthropology 15.
Berkeley: University of California Press.
Kendal, A.
1985 Aspects of Inca architecture: Description, function, and
chronology. BAR International Series 242 (i, ii).
Oxford: British Archaeological Reports.
72 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Kroeber, A. L.
1930 Archaeological explorations in Peru. Pt. II: The northern
coast. Field Museum of Natural History Anthropol-
ogy Memoirs 2:47116. Chicago.
Lozada-Cerna, M.
1987 La ceramica del componente mortuorio de Estu-
quia, Moquegua. Bachelors thesis, Universidad
Catlica Santa Maria, Arequipa.
1998 Seoro of Chiribaya: A bio-archaeological study in
the Osmore Drainage of southern Peru. Ph.D. dis-
sertation, Department of Anthropology, University
of Chicago. University Microlms, Ann Arbor.
Morris, C., and D. E. Thompson
1985 Hunuco Pampa: An Inca city and its hinterland. Lon-
don: Thames and Hudson.
Moseley, M. E.
1992 The Inca and their ancestors. London: Thames and
Hudson Press.
Owen, B.
1993 A model of multi-ethnicity: State collapse, com-
petition, and social complexity from Tiwanaku to
Chiribaya in the Osmore Valley, Peru. Ph.D. dis-
sertation, Department of Anthropology, University
of California, Los Angeles.
Reycraft, R. M.
1997 Calamity, collapse, and continuity: Varied responses
to El Nio in south coastal Peru ca. A.D. 1400.
Paper presented at the 62nd Annual Meeting of the
Society for American Archaeology, April 4, 1997.
Nashville.
1998 The Terminal Chiribaya Project: The archaeol-
ogy of human response to natural disaster in south
coastal Peru. Ph.D. dissertation, Department of
Anthropology, University of New Mexico. Univer-
sity Microlms, Ann Arbor.
Rice, D. S.
1993 Late intermediate period domestic architecture
and residential organization at La Yaral. In Domes-
tic architecture, ethnicity, and complementarity in the
south-central Andes, edited by M. S. Aldenderfer,
6682. Iowa City : University of Iowa Press.
Rice, P. M.
1987 Pottery analysis: A sourcebook. Chicago: University of
Chicago Press.
Sackett, J. R.
1982 Approaches to style in lithic archaeology. Journal of
Anthropological Archaeology 1:59112.
Satterlee, D.
1993 Impact of a fourteenth century El Nio on an indig-
enous population near Ilo, Peru. Ph.D. dissertation,
Department of Anthropology, University of Florida,
Gainesville.
Shennan, S. J.
1989 Introduction. In Archaeological approaches to cultural
identity, edited by S. J. Shennan, 132. London:
Unwin Hyman.
Stanish, C.
1985 Post-Tiwanaku regional economies in the Otora
Valley, southern Peru. Ph.D. dissertation, Depart-
ment of Anthropology, University of Chicago.
1992 Ancient Andean political economy. Austin: University
of Texas Press.
Tschopik, M.
1946 Some notes on the archaeology of the Department of Puno,
Peru. Papers of the Peabody Museum of American
Archaeology and Ethnology 27 (3). Cambridge,
MA: Harvard University Press.
Wiessner, P.
1983 Style and ethnicity in the Kalahari San projectile
point. American Antiquity 48:253276.
1984 Reconsidering the behavioral basis for style: A case
study among the Kalahari San. Journal of Anthropo-
logical Archaeology 3:190234.
Wobst, M.
1977 Stylistic behavior and information exchange. In For
the director: Research essays in honor of the late James B.
Grifn, edited by C. E. Cleland, 317342. Anthro-
pological Papers 6, Museum of Anthropology. Ann
Arbor: University of Michigan Press.
73
INTRODUCTION
The identication of specic groups or individuals in
ancient societies is often, at best, an imperfect task,
fraught with difculties. The principal objective of this
paper is more about presenting problems than propos-
ing denitive solutions about discovering identity or
identities in the archaeological record. This contri-
bution also represents a departure from the general
concept of identity used in archaeology. It usually
refers to the determination of social identity, or the
denition of social and physical boundaries between
different groups on the basis of their material culture.
Our goal is to explore a somewhat more metaphysical
concept of identity as we wish to evaluate the degree(s)
of relationship between Moche iconography and the
physical world of this society. The main concern is how
can we discern and validate the degree of identity of
certain individuals or groups in terms of their social,
symbolic, or even mythical persona? In order to explore
this complex question, we present two test cases drawn
from recent research carried out on the Moche.
Although much work remains to be done on this
subject, it is generally accepted that the Moche was a state-
level society occupying the Peruvian north coast between
the rst and the eighth century. They constructed size-
able ceremonial centers, surrounded in some cases by
well-developed cities with streets and artisanal sectors.
Their funerary practices were complex and elaborate, and
attracted the interest of treasure hunters and collectors for
centuries. The numerous offerings placed to accompany
their dead included among other things metallic objects,
textiles, and nely painted and sculpted ceramics. Tens of
thousands of these beautiful vessels and ceramic objects
have found their way into public and private collections
around the world. The various and complex representa-
tions on these artifacts have been studied by numerous
scholars, and have led to the identication of a number
of individuals involved in various ritual contexts (see for
example: Alva and Donnan 1993; Benson 1972; Donnan
1978; Hocquenghem 1987).
The present investigation will progress along two
different, albeit associated sets of information: the
CHAP T E R 5
Who Were the Priests, the
Warriors, and the Prisoners?
A P E CUL I AR P ROBL E M OF
I DE NT I T Y I N MOCHE CULT URE AND
I CONOGRAP HY, NORT H COAS T
OF P E RU
S TEVE BOURGET
74 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
archaeological record and the iconography found on the
ritual ceramics and murals. In the rst test case, I will
attempt to evaluate the identity of the sacricial victims
found at the recently discovered Huaca de la Luna sac-
ricial site in the Moche Valley (gure 1). Were these
men slain enemies captured during battles? Or were they
part of an elite, shock troops of some sort, sent in sacred
battleelds for the greater good of the community?
To a certain extent, the second test case regards an
even more difcult problem, that is, the very nature
of the Moche rulers, such as those represented in one
of the most complex scenes within Moche painted ico-
nography, the Sacrice Ceremony Theme (gure 2).
In the literature, these seemingly important individu-
als have been irrespectively and often interchangeably
termed priests and priestesses (Alva and Donnan 1993:
132; Hocquenghem and Lyon 1980), gods (Benson
1972:42), or mythical ancestors (Hocquenghem 1987:
125). But how can we assess the nature of their iden-
tity both physical and ideological? How were they
conceptualized in Moche terms? Were these people
perceived as human beings with vested interests in
mundane affairs? Were they perceived as being con-
nected to higher planes of reality, such as a world of
the ancestors and/or the afterlife? It will be argued
that a linear lecture of the evidence within our grasp
does not sufce to dene these actors solely in terms
of their political or social roles, and that more complex
models need to be developed.
HUACA DE LA LUNA SACRIFICIAL SITE
During recent years numerous evidences of human
sacrifices have been found in Moche sites such as
Sipn, Dos Cabezas, and Huaca Cao Viejo (Alva 1994;
Cordy-Collins 2001; Franco, Glvez, and Vsquez
1996). Nonetheless, it is at Huaca de la Luna that the
most elaborate sacricial site was discovered in 1995. It
consisted of more than seventy male individuals sacri-
ced in a special plaza of the temple during numerous
rituals (Bourget 1998, 2001) (gure 3). An open plaza
(Plaza-3A) and a rectangular structure (Platform-II)
bisecting a rocky outcrop form this part of the temple.
Both structures are connected to the north and to the
south by high walls still some eight meters tall in places.
This bipartite construction constitutes a prolongation
of the Huaca de la Luna main platform toward Cerro
Blanco during one of its latest phases of reorganization
(gure 4). On the basis of the new terminology devel-
oped for the construction sequence of the temple, it
would have been associated with the renovation B of
the main platform, A being the very last building to
be erected (Uceda and Canziani 1998). Building A has
been almost completely destroyed by looting activities.
This late extension, Platform-II / Plaza-3A, represents
a single architectural project constructed between the
sixth and the seventh century, during the stylistic phase
Moche-IV.
In Plaza-3A, at least six stratigraphically distinct
sacricial rituals took place in its northwest corner.
This space covering less than 60 m
2
is situated just in
front of the rocky outcrop. Victims embedded in thick
layers of clay indicate that two of these rituals were
performed during spells of torrential rain brought to
this desertic coast by the effects of an El Nio phe-
nomenon. These deposits of clay were most certainly
washed from the high adobe walls surrounding the
plaza. The sacricial sequence and the positioning of
the human remains during each of the rituals, or even
between subsequent rituals, bespeak a high degree of
ritual organization. For example, each sacricial ritual
Figure 1. Human remains in Plaza-3A, Huaca de la Luna.
B O U R G E T 75
Figure 2. Sacrice Ceremony scene, redrawn from Donnan and McClelland 1999.
Figure 3. Plan of Plaza-3A and Platform-II, Huaca de la Luna.
76 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
performed in a layer of clay was followed by a second
one taking place some time after on the dry mud sur-
face. Furthermore, the corpses found embedded in the
clay are usually resting on their backs, whereas the
individuals sacriced during the dry conditions have
been placed facing down right on top of the previous
ones. After each performance the mutilated remains of
the victims were left exposed to the natural elements
and denied a proper burial. Considering these new
archaeological contexts it became obvious that the
numerous scenes of sacrices represented on ceram-
ics, metallic objects, and the walls of the temples may
actually have taken place. The evidence of sacrices
found in each of the Moche sites presently under
investigation also indicates that these practices formed
an integral part of Moche religion and ideology.
All the victims of the sacricial site of Huaca de la
Luna were male, and their age was between 15 and 39
years with an average of 23. On the basis of their age
and sex it thus appears to be a selected group of indi-
viduals. Their skeletal morphology indicates that these
were healthy, strong individuals who were physically
very active. Numerous well-healed fractures on many
of them, and fresh injuries on at least eleven, indicate
that these persons were probably warriors captured dur-
ing violent encounters (Verano 1998, 2001). Although
numerous representations realistically depict scenes of
battle between warriors equipped with protective hel-
mets, backaps, shields, and war clubs, this excavation is
the rst time that what appear to be authentic warriors,
in esh and blood so to speak, were found slain in such
vast numbers in a Moche site.
WAR IN PEACE
Alva and Donnan (1993) have already suggested that the
main objective of these ghts was not to kill the oppo-
nent but to incapacitate him by a blow to the face or to
the legs. By matching together a number of scenes, they
reconstructed a narrative sequence from the moment of
the warriors capture to their arrival at the temples:
Once an enemy was defeated, some or all of his cloth-
ing was removed, a rope was placed around his neck,
and his hands were sometimes tied behind his back.
The prisoners clothing and weapons were made into a
weapon bundle, which was tied to the victors warclub
and slung over his shoulder. The victor held the rope tied
to the prisoners neck and forced the prisoner to walk in
front of him. . . . The prisoners were ultimately taken to
a place where they were formally arraigned before a high
status individual. (Alva and Donnan 1993:131)
Figure 4. Plan of Huaca de la Luna, Moche Valley.
B O U R G E T 77
A number of scenes suggest that these combats
were performed under the guidance or supervision of
high-ranking individuals as well. Such a person is often
portrayed as standing alongside or in the midst of the
battle. On one neline painting (gure 5), the leader is
standing alone with a bird on each side. On the basis
of the attire worn by the warriors, it would appear
that both sides of the warring party belong to Moche
society. Apart from very few examples, the opposing
groups wear similar clothes and carry the same type
of equipment, usually consisting of a conical helmet
or a more elaborate headdress, a backap, a round or
a square shield, and a long club. However, differences
can be found among the designs of the tunics and the
forms of the shields. For example, on the aring bowl
depicted in gure 5, each of the four warriors wears
a stylistically different outt and carries a distinctive
shield. Furthermore, this variation in warfare regalia
on the same representation is not rare but the norm.
It would seem that the objective is indeed to show in a
single scene as many different styles of warriors outts
as possible, and to establish a certain symbolic duality
in the alternation of round and square shields.
Discoveries recently made at two Moche sites sug-
gest that the representations of the warriors small
shields and very long clubs portray the actual objects,
whose depicted sizes compared to human beings are
accurate. At Dos Cabezas, in the Jequetepeque valley,
three small shields made of wood covered with metal
sheet have been found by Christopher Donnan in an
early Moche burial (Donnan 2001). These protective
devices are about 30 cm in diameter. In addition, two
wooden maces were found deposited at the base of a
wall at Huaca Cao Viejo in the Chicama Valley (Franco,
Glvez, and Vsquez 1999). These broken maces, which
have been carved from a single piece of wood, were,
like the shields, originally covered with metal sheet and
measured at least 120 cm.
I would suggest that the small size of the shields
which would have barely protected the forearm and
were probably tied to the left wristand the long,
slender, relatively fragile shafts of the maces would
indicate that indeed these implements must have been
used against opponents similarly equipped in a context
of ritualized warfare. In my point of view, it would have
been suicidal for Moche warriors to engage in regular
warfare against other groups armed only with these war
clubs and shields offering so little protection.
Another problem relating to the use of Moche
iconography to portray scenes of warfare is directly
associated with the very nature of this system of
representation. Numerous scenes represent not only
human beings engaged in battles, but what appear to
be subjects with supernatural attributes as well, such
as animated objects and anthropozoomorphic beings
(gure 6). Moreover, these actors are often engaged
in other contexts involving sacrices, funerary rituals,
marine settings, and so forth. These additional activities
Figure 5. Interior view of a aring bowl. Museum fr Vlkerkunde, Berlin, VA-48171.
78 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
F
i
g
u
r
e
6
.
R
e
v
o
l
t
o
f
t
h
e
O
b
j
e
c
t
s
,
r
e
d
r
a
w
n
f
r
o
m
D
o
n
n
a
n
a
n
d
M
c
C
l
e
l
l
a
n
d
1
9
9
9
.
B O U R G E T 79
suggest that battle scenes were part of a greater concep-
tual whole that must also be explained.
This does not mean that the Moche did not practice
any form of warfare other than ritual. They possibly
did, but it is likely not represented in the iconography
because of its very nature and function. Other contexts
or elements of the material culture will need to be
located in order to demonstrate this important aspect.
Nevertheless, various authors discussing the problem
of secular warfare during the Moche period have often
used these scenes as examples of battles or corroborat-
ing evidence. For example, Wilson (1988) published
the scene on the aring bowl just presented in order
to discuss the practice of Moche warfare in the Santa
Valley (gure 5):
The scene . . . depicts what can probably be considered
as the next stage following the successful conclusion of
a campaign carried out against another region. Nude
prisoners with ropes around their necks are shown
being led across the intervalley desertwith Andean
foothills, coastal birds, Opuntia cactus, and Tillandsia
plants depicted in the background. (1988:338).
The use of this representation to document the exis-
tence of an expansionist state is in fact based on a very
selective use of the iconography. It is not supported by
most of the scenes related to this theme. This highly
selective process is made even more apparent when, in
order to discuss the practice of tribute in the Moche
state, the same author extracts from a complex scene
pertaining to the Burial Theme, a portion showing the
presentation of strombus seashells to a high-ranking
individual seated under the gabled roof of a platform
structure: The bottom scene shows a llama caravan
approaching a ruler with what may be a tribute from a
conquered coastal valley (1988:341). In this case, this
approach shows a complete disregard for the Burial
Theme as a whole, since only a fraction of the depiction
is used to discuss a topic not even remotely connected to
it (see the complete scene in: Donnan and McClelland
1979:16, gure 2). Moreover, strombus seashells can
only be found in the warm coastal waters of Ecuador.
I would suggest that there is no connection between
this scene and tribute exaction. Indeed, Izumi Shimada
(1994) mentions the same problem concerning Moche
warfare in general and warns against the use of these
scenes for this type of reconstitution:
Indeed, purely on the basis of internal analysis of
Mochica neline drawings of combat, any clear secular
motive is not readily forthcoming. . . . The consistent
depictions of captives, including their sacrifice in
association with supernatural beings, provide yet
another line of supportive evidence for the religious
view. Overall, combat scenes are neither convincing
nor conclusive indicators of real-life secular battles for
Mochica territorial expansion. (1994:110)
On the basis of the arguments just presented and the
artifacts recently discovered at Dos Cabezas and Huaca
Cao Viejo, two propositions can be made. First, the
equipment used for secular warfare would not neces-
sarily have been the type that is usually represented in
the iconography. It is likely that more effective weapons
such as spearthrowers and slingshots, which are some-
times represented in other contexts, would have been
used. And second, even if secular warfare existed in
Moche society, which is entirely possible but still dif-
cult to document archaeologically, Moche iconography
is very ritualistic in nature and thus would probably not
have represented this type of activity (Alva and Don-
nan 1993; Donnan 1988; Donnan and Castillo 1994;
Hocquenghem 1987).
Taking into account Moche representations of bat-
tles and the objects found in Moche sites, I would thus
propose that the victims sacriced at Huaca de la Luna
most probably came from the Moche population itself.
The research carried out on the osteological collection
of the plaza by Verano (2001) appears to support the
hypothesis that the sacricial victims represent a very
select group:
Overall, the demographic composition indicates a
highly selective sample of individuals. . . . Moreover,
many of these injuries, especially skull fractures and
broken ribs, are more typical of wounds incurred
through interpersonal violence than from accidents.
Overall, the fracture data suggest that this was a group
with a history of violent encounters. (2001:118)
Their apparent good health, their strengthbased
on the indices of muscular attachmentsand the
numerous ancient injuries suggest that they could have
been part of a group of individuals specically trained
and prepared for these ritual activities. But what about
those who send them to these sacred battleelds?
HUMANS OR ANCESTORS?
Recent archaeological discoveries have made it increas-
ingly apparent that participants in some of the most
important rituals portrayed in the iconography could
have been not only mythological gures, but living
80 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
human beings as well. At Sipn, in the Lambayeque
Valley, Alva (Alva and Donnan 1993) unearthed the
tombs of two high-ranking individuals buried with a
retinue of people and numerous ceremonial artifacts.
On the basis of these objects, such as headdresses, bells,
golden backaps, and scepters, these two males were
eventually identied as the main protagonists, A and
B, of the Sacrice Ceremony (gure 2).
Shortly after, at San Jos de Moro in the Jequete-
peque Valley, Donnan and Luis Jaime Castillo (1992,
1994) discovered the elaborate burials of two women
linked to the same theme and identied as the priestess
C (gure 2). They had been buried with ceremonial
goblets identical to the one exchanged between A and
B, and their cofns had been decorated with metallic
plumes, made of a silver-copper alloy, in many respects
similar to the one depicted in the headdress of individual
C in the Sacrice Ceremony scene. Moreover, in the
tomb of the rst priestess, excavated in 1991, there was
a ceramic goblet adorned with a representation associ-
ated with war and sacrice. A neline painting on the
side of the vessel depicts a procession of animated war
bundles carrying goblets apparently lled with blood
(Alva and Donnan 1993:224, gure 248).
These two burials, dating to the early eighth cen-
tury, were considerably more recent than those of
Sipn, dating to around the fourth century. The cer-
emony apparently was not conned within the limits
of the northern part of the Moche kingdom, for a
mural depicting the same ritual was found in 1958 at
Paamarca in the Nepea Valley, nearly 300 km to the
south (Bonavia 1985:60, gure 39). These ndings con-
rm that the Sacrice Ceremony was enacted all over
the Moche region by high-ranking individuals wearing
similar regalia and using the same paraphernalia during
a period of at least three to four hundred years. Alva and
Donnan (1993) mention that:
The fact that the Sacrice Ceremony was so widespread
in both time and space strongly implies that it was part
of a state religion, with a priesthood in each part of the
kingdom composed of nobles who dressed in prescribed
ritual attire. When members of the priesthood died,
they were buried at the temple where the Sacrice Cer-
emony took place, wearing the same objects they had
used to perform the ritual. Subsequently, other men and
women were chosen to replace them, to dress like them,
and to perform the same ceremonial role. (1993:226)
The reanalysis of a previously excavated tomb situ-
ated more or less halfway between San Jos de Moro
and Paamarca has led to the identication of another
actor of the same ceremony. This tomb, discovered in
1946 at the site of Huaca de la Cruz in the Vir Val-
ley, contained the remains of a woman buried with a
wooden staff. This artifact has permitted Daniel Arse-
nault (1994) to identify her as the participant D who
performs a human sacrice on the lower register of
the Sacrice Ceremony scene. She is portrayed as an
anthropomorphized staff drawing blood from a man
with his hands tied behind his back (gure 2). Interest-
ingly, a second female burial nearby contained a ceramic
goblet along with a jar decorated with two anthropo-
morphized shields and clubs holding similar goblets
(Strong and Evans 1952:141145). These two burials
would date from the Moche-IV stylistic period.
But what are the perceived social identities of these
four actors, given that one of them is portrayed as an
anthropomorphized staff, the principal one has rays
terminated by catsh heads emanating from its body
(A), and the third possesses the head of a nocturnal
bird (B)? For the moment, this is a difcult question
to answer. More research will need to be done on these
supernatural attributes and their associations with spe-
cic individuals. The anthropomorphization of objects
and animals or the corresponding zoomorphization
of human beings is one of the major representational
systems of Moche iconography, alongside ritualized
warfare. Furthermore, ritual battles and humanized
objects and animals are intimately related to one
another not only in the Sacrice Ceremony scene, but
also in the so-called Revolt of the Objects Theme and
elsewhere (gure 6).
THE REVOLT OF THE OBJECTS
Fineline paintings of anthropomorphized objects
capturing human beings were among the rst Moche
representations to be analyzed and published. In 1928,
these scenes were compared by Krickeberg to a Quich
Mayan myth on the rebellion of domestic objects and to
a vaguely similar story collected in the seventeenth cen-
tury in the Peruvian region of Huarochiri (Krickeberg
1928 in Quilter 1990). More recently, this proposition
has been reinvestigated by Jeffrey Quilter (1990, 1996,
1997) in order to demonstrate that effectively these
scenes were part of an ancient myth that spread all
over the Americas. But there exists a fundamental dif-
ference between the revolt of objects as described in the
Popol Vuh, in the Huarochiri manuscript, and depicted
in the Moche representations. It is in the very nature of
B O U R G E T 81
the revolters. The rst two cases relate broadly speak-
ing to culinary implements and domestic animals such
as cooking vessels, grinding stones, dogs, turkeys, and
llamas (Salomon and Urioste 1991:53; Tedlock 1985:
8485). The Moche case consists almost entirely of
warriors implements and ritual objects such as head-
resses, tunics, backaps, lances, spearthrowers, shields,
and nose ornaments. Quilter (1990) has also noted this
major difference and he associated it with a local varia-
tion of a pan-American myth:
That items associated with military costume and activi-
ties are shown in revolt instead of metates and llamas
suggests that the Moche took an ancient story and
created their own particular variant which was rein-
terpreted by later peoples, such as in the ethnohistoric
tales. (Quilter 1990:50)
I consider that the use of war and ritual implements
instead of culinary implements or domesticated animals
is an extremely important variant, and that it must
be taken into account in order to explain some of the
meaning of these peculiar scenes. In fact, I would argue
that those animated weapons and military ornaments
capturing human beings, often denuded, ts well within
the logic of Moche ritual warfare as reiterated in numer-
ous occasions in the iconography. In some, lhabit fait le
moine, and a man vested with these attributes becomes
a fantastic warrior, while another individual deprived of
them by force becomes a captive and eventually a vic-
tim of sacrice. The iconography abounds with scenes
representing warriors ghting each other, usually in
some form of duel, and more rarely in a group. The
defeated are then denuded, completely or partially, and
their outt and weapons are secured to the war maces of
the victorious. The attributes of the defeated are always
clearly represented and seem to be as important as the
rest of the depiction (gure 5).
The Munich vase, which is from the Moche-IV
period and portrays the most complex rendition of the
Revolt of the Objects, features the four actors encoun-
tered in the Sacrice Ceremony scene and recognized
in the burials at Sipn, San Jos de Moro, and Huaca
de la Cruz (gures 7, 8). I would suggest that this scene
might represent the event that will eventually lead to
the Sacrice Ceremony. In accord with Quilter (1990),
who published the most extensive analysis of this scene,
I also think that it portrays a narrative involving two
distinct groupings of actors. But I would like to suggest
an alternative proposition to his idea that it represents a
revolt by the actors of the lower register, brought under
control by the rayed deity of the upper register (gure
7, A): The defeat of the revolting objects must be
assumed to represent generally the establishment of
order as opposed to the chaos of the dark forces that
caused the uprising (Quilter 1990:53). I will attempt
to demonstrate that it is a narrative of ritual warfare
and capture that ts well with the general ideology
expressed in the iconography.
On the upper register are actors A and C super-
vising a group of anthropomorphized animals (gure 7).
These assistants are well-known entities and are often
depicted as warriors, as in this case; messengers; and in
some cases, sacricers. To the left and to the right, the
terrestrial animals are dragging a musical instrument, a
conical helmet, and two nose ornaments toward actors
A and C (gure 6). In the center, just below the
couple, three avian beings are respectively ghting a
warriors tunic, a shield, and a serpent-fox belt. I would
propose that these scenes represent the rst part of the
narrative, with A and C sending their anthropo-
zoomorphic helpers to bring, by force or by will, ritual
implements to take part in a sacred battle.
On the lower register of this neline painting (gure
8), humanized implements are now defeating and captur-
ing human warriors and an anthropomorphic deer. This
battle takes place under the supervision and guidance of
the bird impersonator B and woman C, who also
participates in the upper register. Her participation in
both sections reinforces the hypothesis that this series
represents a narrative. In this complex battle scene,
human warriors are captured, stripped of their clothing,
and taken prisoner under the watchful eyes of A, B,
and C of the Sacrice Ceremony scene. Furthermore,
the characteristic humanized staff with four prongs, iden-
tied as sacricer D in the Sacrice Ceremony scene,
is also involved in the taking of prisoners. To the right
of the Revolt of the Objects scene, the staff is depicted
dragging a warrior by the hair (gure 8).
I think that Quilter is right to propose that this
scene will culminate with the eventual sacrice of the
captives during the Sacrice Ceremony (Quilter 1990:
55). In this scene, all the categories of actors are also
represented: human beings, zoomorphic beings, and
animated objects (gure 2). I consider that this scene
could also be a two-step narrative. In the rst instance,
on the lower register, an anthropomorphized feline
and the humanized staff with four prongs, individual
D, are now carrying out the sacrices. Both are rep-
resented with their corresponding weapon bundles. In
the following phase of this ritual, on the upper register,
82 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
F
i
g
u
r
e
7
.
R
e
v
o
l
t
o
f
t
h
e
O
b
j
e
c
t
s
,
d
e
t
a
i
l
r
e
d
r
a
w
n
f
r
o
m
D
o
n
n
a
n
a
n
d
M
c
C
l
e
l
l
a
n
d
1
9
9
9
.
F
i
g
u
r
e
8
.
R
e
v
o
l
t
o
f
t
h
e
O
b
j
e
c
t
s
,
d
e
t
a
i
l
r
e
d
r
a
w
n
f
r
o
m
D
o
n
n
a
n
a
n
d
M
c
C
l
e
l
l
a
n
d
1
9
9
9
.
B O U R G E T 83
A and B are partaking human blood brought by
woman C in a special container.
Animated warrior attributes capturing prisoners could
well reinforce the concept of ritualized warfare within the
same population. For individuals originating from the
same social group but ghting each other, these objects
could have symbolized the ritual activity itself. In the
Revolt of the Objects scene, the paramount warlord A
forces the objects to obey his orders, while B and C,
standing on the summit of a temple, receive the captives.
Thus an individual clothed and armed with these almost
magical attributes becomes a fantastic warrior acting
under the strict guidance of A, B, and C, and
likewise, someone deprived of these objects becomes a
prisoner and eventually a sacricial victim.
POLITICS AND IDEOLOGY
The theme of the Revolt of the Objects does not appear
suddenly during the late Moche-V period, but exists
during the Moche-IV period as well. As mentioned, the
representations of anthropomorphized war implements
and objects during Moche-IV are numerous. The main
difference is that during the later Moche-V, the theme
is expressed in its most complex form. This increase in
pictorial complexity is found in the Burial Theme and
other themes as well during this period (Donnan and
McClelland 1999).
Anthropomorphized implements would have sig-
nied the will of Moches highest order and, for the
population or the people involved in these battles, it
could have reduced the tensions generated by this activ-
ity. Indeed, many scenes give equal importance to the
warriors, the prisoners, or the war implements. The
frequent representations of war bundles, and in some
cases of people literally worshipping these objects, dem-
onstrate their paramount importance. Like the animals
represented in the iconography, these are not simple
objects, but subjects in their own right, who take an
active role in the construction of the ideology transmit-
ted by this system of representation.
I would suggest that this scene does not depict a
revolt against a pre-established order but the very reit-
eration of this sacred order. The objects involved are
war or ritual implements fullling their very function.
The victims are not simple individuals but full-edged
warriors. It might even be more appropriate to rename
this complex scene the Battle of the Objects instead.
With this representation of animated objects, we are
not confronted with a semantic shift but with the dem-
onstration of the importance of ritualized warfare in
Moche symbolism, religion, and ideology.
In the case of the Huaca de la Luna sacricial site,
these ritual battles would have been performed, among
other things, for restoring proper ecological conditions
(Bourget 1998, 2001). The ancestors or entities con-
trolling El Nio phenomenon chose the victims on the
sacred battleeld itself. Their will and their choices are
made apparent by the defeat of some of the champions.
The victorious warriors then strip naked the defeated
and bring them in a procession to the temple. They are
paraded inside the building in front of murals portray-
ing fantastic beings such as catshes and octopuses,
creatures intimately linked with the world of the sea and
possibly El Nio events. Eventually the chosen ones,
though the ordeal of ritual battles, are prepared and
sacriced within the most private part of the temple.
On the basis of the burials found so far at a number
of Moche sites and the variety of actors represented in
the iconography, it appears that the Moche priesthood
and/or ruling elite was complex and composed of many
different types of ofciates. If the principal actors of
the Sacrice Ceremony scene could also have been
portrayed by living individuals, such as those buried at
Sipn, San Jos de Moro, and Huaca de la Cruz, it is
plausible that their identity was closely associated with
a certain world of the ancestors.
The very nature of the Moche political and religious
system is an important issue, and more research will
need to be done. The symbolically charged content of
Moche iconography, in regard to ritual warfare or the
representations of high-ranking individuals, signals a
complex system of relations between political rulers,
ceremonial practices, and religious beliefs. The rulers
of the Sacrice Ceremony scene seem to be part of an
elite that transgresses the boundaries between life and
death, between the world of the living and the world
of the ancestors. Did the Moche population perceive
some of their rulers as living ancestors? It is possible
this would have justied, in an almost naturalor in
this case supernaturalway, their right to rule.
This phenomenon is more than a mere manipulation
of the masses. The dialectics between the population and
these possible living ancestors, or other entities closely
associated with the onset of cataclysmic events, must
have been extremely elaborated. This important aspect
is still under investigation. By recognizing some possible
links between Moche rulership and a more metaphysical
political system, we are only scratching the surface of a
very complex and exciting problem indeed.
84 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
REFERENCES
Alva, W.
1994 Sipn. Edited by J. A. de Lavalle. Coleccin Cultura
y Artes del Peru. Lima: Cerveceria Backus & John-
ston S.A.
Alva, W., and C. B. Donnan
1993 Royal tombs of Sipn. Exh. cat. Los Angeles: Fowler
Museum of Cultural History, University of
California.
Arsenault, D.
1994 Symbolisme, rapports sociaux et pouvoir dans
les contextes sacrificiels de la socit Mochica
(Prou prcolombien). Une tude archologique et
iconographique. Ph.D. dissertation, Department of
Anthropology, University of Montreal.
Benson, E. P.
1972 The Mochica: A culture of Peru. London: Thames and
Hudson.
Bonavia, D.
1985 Mural painting in ancient Peru. Translated by P. J.
Lyon. Bloomington: Indiana University Press.
Bourget, S.
1998 Pratiques sacricielles et funraires au site Moche
de la Huaca de la Luna, cte nord du Prou. Bul-
letin de lInstitut Franais dEtudes Andines 27 (1):
4174.
2001 Rituals of sacrice: Its practice at Huaca de la Luna
and its representation in Moche iconography. In
Moche art and archaeology in ancient Peru, edited by
J. Pillsbury, 89109. Washington, D.C., and New
Haven: National Gallery of Art and Yale University
Press.
Cordy-Collins, A.
2001 Decapitation in Cupisnique and early Moche societ-
ies. In Ritual sacrice in ancient Peru: New discoveries
and interpretations, edited by E. P. Benson and A.
Cook, 2134. Austin: University of Texas Press.
Donnan, C. B.
1978 Moche art of Peru: Pre-Columbian symbolic communi-
cation. Exh. cat. Los Angeles: Museum of Cultural
History, University of California.
1988 Unraveling the mystery of the warrior-priest.
National Geographic 174 (4): 550555.
2001 Moche burials uncovered. National Geographic 199
(3): 5873.
Donnan, C. B., and L. J. Castillo
1992 Finding the tomb of a Moche priestess. Archaeology
45 (6): 3842.
1994 Excavaciones de tumbas de sacerdotisas Moche en
San Jos de Moro, Jequetepeque. In Moche: Propues-
tas y perspectivas, edited by S. Uceda and E. Mujica,
415424. Actas del Primer Coloquio sobre la Cultura
Moche (Trujillo, 12 al 16 de abril de 1993). Travaux
de lInstitut Franais dEtudes Andines 79. Lima.
Donnan, C. B., and D. McClelland
1979 The burial theme in Moche iconography. Studies in
Pre-Columbian Art and Archaeology 21. Washing-
ton, D.C.: Dumbarton Oaks Research Library and
Collection.
1999 Moche neline painting: Its evolution and its artists.
Los Angeles: UCLA Fowler Museum of Cultural
History.
Franco, R., C. Glvez, and S. Vsquez
1996 Los descubrimientos arqueolgicos en la Huaca Cao
Viejo. Arkinka 1 (5): 8294.
1999 Porras Mochicas del complejo El Brujo. Revista
Arqueolgica SIAN 4 (7): 1623.
Hocquenghem, A. M.
1987 Iconografa Mochica. Lima: Ponticia Universidad
Catlica del Per.
Hocquenghem, A. M., and P. Lyon
1980 A class of anthropomorphic supernatural females in
Moche iconography. awpa Pacha 18:2748.
Krickeberg, W.
1928 Mexikanisch-Peruanische parallelen, Ein Uber-
blick und eine Erganzung. In Festschrift: Publication
dhommage offerte au P.W. Schmidt, edited by W.
Koppers, 378393. Vienna: Congregations-Buch-
druckerei-Mechitharisten.
Quilter, J.
1990 The Moche revolt of the objects. Latin American
Antiquity 1 (1): 4265.
1996 Continuity and disjunction in pre-Columbian art
and culture. RES 2930:303317.
1997 The narrative approach to Moche iconography.
Latin American Antiquity 8 (2): 113133.
Salomon, F., and G. L. Urioste, translators
1991 The Huarochir manuscript: A testament of ancient and
colonial Andean religion. Austin: University of Texas
Press.
Shimada, I.
1994 Pampa Grande and the Mochica culture. Austin: Uni-
versity of Texas Press.
Strong, W. D., and C. Evans, Jr.
1952 Cultural stratigraphy in the Vir Valley, northern Peru.
Columbia Studies in Archaeology and Ethnology 4.
New York: Columbia University Press.
Tedlock, D.
1985 Popol Vuh: The Mayan book of the dawn of life. New
York: A Touchstone Book, Simon & Schuster.
B O U R G E T 85
Uceda, S., and J. Canziani
1998 Anlisis de la secuencia arquitectnica y nuevas per-
spectivas de investigacin en la Huaca de la Luna.
In Investigaciones en la Huaca de la Luna 1996, edited
by S. Uceda, E. Mujica, and R. Morales, 139158.
Trujillo: Facultad de Ciencias Sociales, Universidad
de la Libertad.
Verano, J.
1998 Sacrificios humanos, desmembramientos, y
modicaciones culturales en restos arqueolgicos:
Evidencias de las temporadas de investigacin
199596 en Huaca de la Luna. In Investigaciones en
la Huaca de la Luna 1996, edited by S. Uceda, E.
Mujica, and R. Morales, 159171. Trujillo: Facultad
de Ciencias Sociales, Universidad de la Libertad.
2001 War and death in the Moche world: Osteological
evidence and visual discourse. In Moche art and
archaeology in ancient Peru, edited by J. Pillsbury,
111125. Washington, D.C., and New Haven:
National Gallery of Art and Yale University Press.
Wilson, D.
1988 Prehispanic settlement patterns in the lower Santa Valley
Peru: A regional perspective on the origins and develop-
ment of complex north coast society. Washington, D.C.:
Smithsonian Institution Press.
86
INTRODUCTION
Household approaches to the study of prehispanic
Andean societies have received increasing attention
in the last two decades of archaeological research, as
projects in the central Andes guided by the theoreti-
cal and methodological underpinnings of household
archaeology have been conducted in recent years
(Aldenderfer 1993; Bawden 1982; Bermann 1994, 1997;
Bermann and Castillo 1995; Goldstein 1993; Rice 1993;
Stanish 1992; Stanish, de la Vega, and Frye 1993; van
Gijseghem 2001). These research programs have dem-
onstrated the utility of a household-based approach to
the study of Andean prehistory. Moreover, these stud-
ies have shown that an integrated methodology that
evaluates both domestic architecture and associated
domestic artifact assemblages is necessary not only for
a successful household archaeology (Stanish, de la Vega,
and Frye 1993:83), but for a successful archaeological
measure of ethnicity in the domestic context (Aldender-
fer and Stanish 1993; Stanish 1989, 1992; Stanish, de la
Vega, and Frye 1993).
Archaeological treatises of ethnicity dene the con cept
loosely as all those social and psychological phenomena
associated with a culturally constructed group identity
(Jones 1997:xii). That an ethnic group has a membership
which identies itself (Barth 1969:11) is usually recog-
nized as the dening element of ethnicity as opposed to
other aspects of social identity, a point recently reiter-
ated in several case studies and reviews of ethnicity in
the archaeological record (Aldenderfer and Stanish 1993;
Emberling 1997:302; Wells 1998).
In this paper, I follow Aldenderfer and Stanish (1993)
and Stanish (1989, 1992) by suggesting that domestic
architecture can be used by archaeologists as a measure
of ethnic identity. By employing a household archaeol-
ogy approach, I present the domestic architecture of
Early Nasca society from the southern Nasca region
and demonstrate that this component of residential sites
in the region is (1) an independent measure of ethnicity,
and when compared spatially and temporally to other
sites in the region (2) reveals specic socioeconomic
patterns that are characteristic of Early Nasca society.
CHAP T E R 6
Household Approaches to
Ethnicity on the South
Coast of Peru:
T HE DOME S T I C ARCHI T E CT URE
OF E ARLY NAS CA S OCI E T Y
KEVI N J . VAUGHN
V A U G H N 87
To that end, the domestic architecture and the asso-
ciated artifact assemblages of the Early Nasca site
Marcaya are presented. The architecture of Marcaya
is then compared to the architecture of contempora-
neous domestic sites in the region that exhibit similar
architectural patterns. It is argued that, at least in the
southern Nasca region, Early Nasca domestic archi-
tecture is diagnostic of the ethnicity of the local group.
This conclusion is then used to evaluate changing
socioeconomic and ceremonial patterns in the Nasca
region over time.
ARCHAEOLOGICAL CONTEXT
The prehispanic Nasca culturefamous for exquisite
polychrome neware, geoglyphs (the Nasca Lines),
and the ceremonial/pilgrimage center Cahuachi
developed in the Ica-Grande region, comprising the
Ica Valley and the Rio Grande de Nasca drainage sys-
tems (Carmichael 1998:214; Schreiber 1999; Silverman
1996; Vaughn 2000; gure 1). The entire region is very
dry with infrequent rainfall (ONERN 1971). The riv-
ers that run through the valleys in the region are only
intermittently lled with water (ONERN 1971) and are
thus classied as inuent streams (Schreiber and Lan-
cho Rojas 1995). While the Nasca culture developed in
the Ica Valley (DeLeonardis 2000; Massey 1986), the
northern Nasca region including Palpa (Browne 1992;
Browne and Baraybar 1988; Reindel and Isla 1998), and
the Ingenio Valley (Silverman 1993a), the primary focus
of this paper is what I refer to as the southern Nasca
region (hereafter SNR) including the four southern
tributaries of the Rio Grande (Schreiber 1998, 1999;
Schreiber and Lancho Rojas 1995; gure 1).
0 50 km
N
T
ie
r
r
a
s
B
la
n
c
a
s
T
a
ru
g
a
L
a
s
T
ra
n
c
a
s
A
ja
Cahuachi
Marcaya
Ica
Ro Grande
Palpa
Ingenio
Nasca
Figure 1. Map of the Ica-Grande region of the south coast of Peru with sites mentioned in text. The gray box delineates the
southern Nasca region (SNR).
88 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
The development of Nasca society has been thor-
oughly summarized elsewhere (Carmichael 1988,
1998; Clarkson 1990; DeLeonardis 2000; Schreiber
1998, 1999; Schreiber and Lancho Rojas 1995; Sil-
verman 1993b, 1995; Vaughn 2000). In short, during
the span of the Early Intermediate period (circa AD
1750) Nasca culture developed, reached a orescence,
and was eventually conquered by the Wari Empire in
the beginning of the Middle Horizon (AD 750; Carmi-
chael 1988, 1998; Conlee 2000; Schreiber 1998, 1999;
Schreiber and Lancho Rojas 1995). The emphasis in
this paper is on Early Nasca society (Carmichael 1998:
4; Schreiber 1998: table A-1; Schreiber and Lancho
Rojas 1995: table 2).
Most scholars agree that during its height Nasca
was a confederacy of chiefdoms (Silverman 1993b)
with little sociopolitical stratification (Carmichael
1988, 1995; Silverman 1993b; though see Reindel and
Isla 1998), integrated by the ceremonial center Cahua-
chi (Oreci 1993; Silverman 1993b, 1994a). Helaine
Silverman has maintained that instead of serving as a
sociopolitical center of the Nasca polity during Early
Nasca times, Cahuachi was a ceremonial center where
people from around the south coast of Peru made peri-
odic pilgrimages and built monumental constructions.
The majority of people in the SNR lived in the foot-
hills of the Andes in small, relatively undifferentiated
habitations. Indeed, in the SNR dominated by Cahua-
chi, Schreiber (1990, 1994, 1998, 1999; Schreiber and
Isla Cuadrado 1996; Schreiber and Lancho Rojas 1995)
has located over ve hundred habitation sites through
full-coverage settlement survey, approximately fty of
which date to the Early Nasca period/culture. Early
Nasca habitation sites are located in the foothills just
above the dry coastal plain. The sites are located on
the hillsides anking the valley oor, as prehispanic
occupants of the area utilized the valley bottoms for
agriculture. The sites are small, generally less than
three hectares in size, and do not have evidence of
monumental or ceremonial architecture on the sur-
face (Schreiber 1999; Schreiber and Isla Cuadrado
1996). The eldwork at Marcaya was used to evalu-
ate questions of Early Nasca village economy and to
record the nature of Early Nasca domestic architecture
(Vaughn 2000). The research program was guided by
assumptions of a household archaeology approach (e.g.,
Stanish 1989, 1992). Insofar as this was the primary
research directive, that methodological approach is
summarized briey here.
Table 1. Peruvian and south coast chronology. For simplication, detail is only given for the Nasca culture
(after Carmichael 1998; Conlee 2000; Schreiber 1998: table A-1; Schreiber and Lancho Rojas 1995: table 2;
Vaughn 2000: table 1.1).
Horizons and
Intermediate
Periods
Culture names Nasca phases Approximate dates
Late Horizon Inka AD 14761532
Late Intermediate
Period
Tiza AD 10001476
Middle Horizon Wari Loro 8 AD 7501000
Early Intermediate
Period
Late Nasca 6, 7 AD 550750
Middle Nasca 5 AD 450550
Early Nasca 2, 3, 4 AD 1450
Proto Nasca 1 100 BCAD 1
Early Horizon Paracas 800100 BC
Initial Period 1800800 BC
Archaic 90001800 BC
Paleoindian 12000?9000 BC
V A U G H N 89
HOUSEHOLD ARCHAEOLOGY
Though households were employed as a primary ana-
lytical unit in Andean research a few decades ago (see
Bawden 1982 for an excellent example), it was not until
the late 1980s that household archaeology became a
commonly used research methodology in Andean
archaeology. The increase in interest was created in
part by several publications by Stanish (1989, 1992),
who imparted a dissatisfaction with previous attempts to
test models of zonal complementarity in the Andes (e.g.,
Murra 1972, 1985). Previous efforts to evaluate ethnic
enclaves in the south-central Andes had centered on
artifact-based methodologies that relied upon stylistic
comparisons of (predominantly) ceramic newares from
tombs (Stanish 1989:8). Mortuary contexts, however,
were argued to be a misleading indicator of ethnicity
because by their very nature, (they) represent areas in
which everyday activities are not performed (Stanish
1989:12). As an alternative, it was argued that archae-
ologists should look to domestic architecture and
associated domestic artifact assemblages to evaluate
the ethnicity of a residential group. The argument was
that these assemblages are much more characteristic of
the residential group because domestic areas tend to be
more conservative in the quantity and variety of exotics
present when compared to mortuary contexts (Stanish
1992:33). This model was found to be an effective one
in identifying prehispanic ethnic groups, particularly
in the south-central Andes (see the contributions in
Aldenderfer 1993).
The archaeological household was argued to be not
only the most appropriate analytical method for den-
ing ethnically distinct populations, it was also a useful
tool for evaluating economic and political relationships
within archaeological settlements (Stanish 1989:7). The
term archaeological household was used to distinguish
between the ethnological concept of household and a
modied concept suitable for archaeological analysis.
Citing Bender (1967), Stanish lists three components
that are essential to the concept of the ethnological
household: coresidence, domestic functions, and familial
relations (Stanish 1989:8). Of these three, familial rela-
tions will not be accessible to the archaeologist, so the
focus for distinguishing the archaeological household
is on coresidentiality and domestic functions. Thus,
the archaeological household is dened as the minimal
coresidential domestic group (Stanish 1989:11).
Minimum coresidential domestic groups should have
the following: 1) there should be spatial segregation of
individual structures or structure groups at a particular
archaeological settlement; 2) these structures or structure
groups should contain material correlates of domestic
activities such as storage and food preparation; and 3)
these basic structure groups and material correlates
should be repeated within a community. It is through the
combination of these three features, evaluated through
surface analysis (when applicable) and excavations, that
archaeological households can be detected.
Households at Marcaya
This methodological approach has wide applicability
and can be employed in other regions of the Andes to
evaluate the ethnicity of a local population. Previous
research in the SNR has suggested that Early Nasca
sites are composed of structures with distinct architec-
ture that is characterized by agglutinated and dispersed
round structure groups situated on the hillsides of the
upper valleys (Schreiber 1988, 1990; Schreiber and Isla
Cuadrado 1996). Marcaya, located on the northern hill-
side of the Tierras Blancas River Valley at 1,000 meters
above sea level, is one such residential site (gure 1).
A research strategy directed by the tenets of house-
hold archaeology should be able to identify and evaluate
archaeological households within the remains of the
prehispanic community of Marcaya. Furthermore, once
archaeological households are identied, the domestic
architecture and the associated domestic artifact assem-
blages can be considered diagnostic of the ethnic group
in residence there. Once presented, these attributes of
the ethnic group are useful for comparison with other
Early Nasca residential communities in the region.
Marcaya is composed of over seventy isolated and
agglutinated structures (gure 2). Since there is little
deposition in this environment, architectural foundations
are exposed on the surface, making it easy to document
the spatial layout of the site. Initial surface inspection at
Marcaya suggested that there were two basic structure
types on the surface, houses and patios, which can be
distinguished on the basis of their material remains,
archaeological features, and spatial organization.
Houses are round and relatively small, with an aver-
age diameter between three and ve meters. The large
quantities of wall fall surrounding the original founda-
tions indicate that walls were originally one and a half
to two meters in height. Houses were constructed with
unmodied to slightly modied eldstone set in mud
mortar, with well-dened interior wall faces. Partially
dressed fieldstone was placed with their modified
surfaces on the inside of structures, creating a clearly
90 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
delineated interior space (gure 3). One or two doors
in houses were placed in the walls that lead either to the
exterior of the house or to associated patios (described
below). Doors are discernible in the architecture, as they
are formed by a seventy-centimeter to one-meter gap
in wall construction and emphasized with two modied
vertical stones placed within the gap (gure 4).
Fallen pilasters measuring up to one meter were
recorded in the interiors of some of the houses. Simple
roofs of brush and branches in the houses were either
lain over the pilasters, or, when pilasters were not
present, roofs could have been supported over wall
foundations to provide rudimentary protection from the
wind, cold, sun, and perhaps more rarely, rain. Roofs of
a similar nature exist on modern dwellings located in
the river valley downstream (gure 5).
Most houses were adjoined by a patio on one or more
sides. Patios are larger and ovoid to rectangular with a
length along the longest axis between six and fourteen
meters. Analysis of abutments demonstrates that patios
Figure 2. Topographic map of Marcaya with architectural features.
V A U G H N 91
Figure 3. Typical patio group at Marcaya (Patio Group V). The house (Structure 12, in the center) is anked by two patios on either side (Structures
11, to the right/west, and 13, to the left/east).
Figure 4. Typical house door at Marcaya. This door leads from Structure 12 to Structure 13 of Patio
Group V. Illustration by Dawn Vaughn.
Figure 5. Modern residential structure located in the Tierras Blancas River Valley approximately one kilometer
downstream (west) from Marcaya.
92 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
were constructed after houses (gure 6). The walls of
patios were apparently much less substantial, as there is
little wall fall surrounding the foundations. Patios were
constructed by clearing the area adjacent to a house of
debris and rock and lining the perimeter of the cleared
area with unmodied eldstone to dene an ovoid to
rectangular space. Retaining walls were usually built
on the southern perimeter of the structure, the side of
the structure that faced downslope. Internal walls were
rare in patios.
Patio Groups
The predominant architectural pattern at Marcaya
is the cluster of houses and patios. There are at least
twenty-three such clusters that can be distinguished
on the surface of the site (gure 7). In addition to these
clusters, there are thirteen structures not associated
with obvious patios. Excavations in these structures
revealed that some were dwellings without adjoining
patios, while the functions of others remain undeter-
mined. Some of the larger ones, especially in Locus 2,
may have been camelid corrals; however, this remains
to be veried.
Based on the methods of dening archaeological
households outlined above, these structure groups
appeared to follow the rst and third requirements set
out for an archaeological household; they are spatially
segregated (i.e., the clusters are distinct) and the groups
are repeated within the community. Thus, the clusters
of structures were referred to as patio groups because
they invariably consist of at least a single house associ-
ated with one or more patios.
In addition to patio groups appearing spatially to
be archaeological households, excavations revealed
artifacts and features that support this proposition.
Excavations were concentrated in ve patio groups,
while others were tested with trenches (Vaughn 2000;
gure 7). No midden contexts were found in excava-
tions, as the depositions in certain contexts have been
deated. It is also possible that inhabitants of the com-
munity were disposing of garbage over the side of the
hill into the river valley below, where these remains
would have since been subjected to aeolian processes,
which can be severe in the narrow valley. Nevertheless,
the remains within the structures themselves conrm
that they were domestic and functioned as residential
loci. Evidence of domestic activities found within the
Figure 6. Planview of the construction sequence in Patio Group V showing the abutment of a patio (Structure 13) to a
house (Structure 12). Illustration by Dawn Vaughn.
V A U G H N 93
Figure 7. Schematic of patio groups at Marcaya. For ease of presentation, patio groups above the dotted line are not drawn to
scale. Excavated structures include any structure with more than half of its units excavated (see Vaughn 2000).
94 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
patio groups included artifacts and features relating to
the processing, preparation, and storing of comestibles,
and small-scale lithic and textile production.
Artifacts and Features
Evidence of food processing includes a variety of
ground stone tools such as large grinding stones
(batanes) and fragments of mortars and pestles. Sub-
sistence goods were stored in subsurface, rock-lined
pits measuring about fty centimeters to one meter in
diameter and one to one-and-a-half meters deep (g-
ure 8). These are similar to features called collomas,
used for storage of foodstuffs and found in modern
highland communities such as Chuquimaran, located
approximately forty kilometers from Marcaya. Such
features are also reported in the Chincha Valley at
Huaca Santa Rosa; Kroeber (1944:137) originally
recorded them as tombs, but subsequent researchers
(Massey 1986:315; Wallace 1971) suggested that they
were storage pits.
Based on their similarity in appearance to mod-
ern-day features, the prehistoric pits were given the
name colloma. At Marcaya, some collomas were found
empty, some lled with sterile sand, and others contain-
ing domestic garbage consisting of potsherds, faunal
material, broken spindle whorls, and food remains.
Based on comparison with the modern features and
their contents, it is suggested that the collomas found
at Marcaya were most likely used for domestic food
storage. Probably due to a combination of a lack of soil
formation and extensive wind erosion in this location,
some collomas are visible on the surface today, while
others were encountered during excavations. It was
found that there is at least one and sometimes more
than one in each dened patio group.
Evidence of the preparation of food was found in
patios, as hearths were present either as formal archi-
tectural features or as ash stains found in excavations
(gure 9). Moreover, fragments of cookpots (ollas) as
well as entire vessels were recovered in excavations.
Due to poor preservation conditions and the lack of
stratied middens, food remains were relatively rare.
Remains that were recovered, however, consisted of
small maize cobs, faunal material from at least four
terrestrial genera, and shellsh remains from six gen-
era (Vaughn 2000).
Evidence for activities other than those relating to
food production and consumption was also present in
excavated patio groups. Signs of lithic productionin
the form of debitage and associated formal tools of
obsidian and chalcedony, and chert and basalt deb-
itagewere found in all patio groups, though a high
concentration of these materials was found in Structure
11 (Vaughn 2000; Vaughn and Glascock i.p.). Spindle
whorls were present in all patio groups as well, indicat-
ing household textile production (Conlee and Vaughn
1999; Vaughn 2000).
Figure 8. Two adjacent collomas found in Structure 27.
V A U G H N 95
Figure 9. Schematics of two typical patio groups (V and XII) with artifact concentrations and features found in excavations. A full range of
domestic remains was found in excavations at Marcaya.
96 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
The vessel assemblage at Marcaya, although com-
posed of a substantial amount of domestic wares such as
ollas and storage jars, also consisted of a high percent-
age of Nasca polychrome neware. Nasca neware is
often regarded as the hallmark of Nasca society (Proulx
1983), and these nely made ceramics were at least in
part responsible for Nascas reputation in Western aca-
demia as an ancient Andean civilization. In previous
research it has been noted that the ceramics, which are
pre-re slip-painted with up to fteen distinct colors
(Carmichael 1998; Proulx 1968), were ubiquitous in
Nasca society, as they were present in mortuary, cere-
monial, and presumably domestic contexts (Carmichael
1988, 1995:231, 1998:216; Silverman 1993b:339). The
proposition that domestic contexts contained a great
deal of neware was based on the observation that poly-
chrome ceramics covered the surfaces of domestic sites
recorded in surface survey, even though excavations had
yet to be conducted in these contexts (Carmichael 1998:
222; Silverman 1993b).
Following Silverman (1993b:228), the number of
fineware vessels (defined as a painted, polychrome
vessel) in the overall vessel assemblage was compared
to the number of plainware (non-painted) vessels.
Fineware vessels included bowls, vases, dishes, head
jars, and painted storage jars (Proulx 1968). Plainware
vessels consisted entirely of large utilitarian collared
vessels that can conveniently be classied as ollas and
jars. The frequencies of vessel types were evaluated with
a Minimum Number of Individual (MNI) calculation
using non-conjoining rim sherds as a proxy measure of
entire vessels. The results of the analysis demonstrate
that there was a high proportion of neware in service
at Marcaya (table 2).
The neware vessels probably served a variety of
functional purposes, as they are the only vessel type
(bowls and vases) in the assemblage adequate for serv-
ing food and drink (Vaughn 2000). As the primary
purveyor of ideology (Carmichael 1998) in Nasca
society, however, they also served a ritual purpose that
involved fertility and water rites (Carmichael 1992a,
1994). Although the neware vessels present at Mar-
caya were apparently not manufactured within the
community (Vaughn 2000; Vaughn and Neff 2000),
as an integral part of the residential vessel assemblage
they are, however, indicators of the ethnicity of the
local group. This is somewhat surprising in compari-
son to examples from the south-central Andes, where
exotics and artifacts of high symbolic and economic
value that were not manufactured by the local com-
munity are generally poor reections of that ethnic
group (Stanish 1992:30; though see Reycraft 1998).
On the contrary, even though the presence of neware
would initially appear to misrepresent the identity of
the local group, its frequent use in domestic contexts
at Marcaya demonstrates that the presence of Nasca
polychrome neware is, in fact, a dening character-
istic of Nasca ethnicity.
All excavated patio groups had the composite of these
material correlates to domestic activities. Cooking areas
were found in each patio group, as were features for
storing and processing food, and the assemblage in each
group consisted of a full suite of domestic artifacts. Thus,
excavations conrmed that those patio groups dened in
surface analysis were indeed archaeological households.
Beyond identifying the characteristics of the local
group, the household archaeology approach facilitates
analysis of the economic and political relationships
within settlements. The principal nding at Marcaya
in this regard is that households were economically
independent of one another. That is, if features and
artifact assemblages are taken as an indication, house-
holds processed, cooked, and stored their own food and
water and were independently producing both textiles
and lithics. Evidence of community organization of
these activities is absent in the analysis of Marcaya.
Furthermore, while neware ceramics were used in
relatively mundane contexts involving serving and con-
suming food and drink, they are also heavily laden with
symbolic iconography, and thus, were undoubtedly used
Table 2. Comparison of the vessel assemblage at Marcaya and Cahuachi.
Marcaya Cahuachi (Silverman 1993a:228)
# % # %
Plainware Sherds 9253 88 3532 78
Fineware Sherds 1269 12 999 22
Plainware Rims 170 44 140 29
Fineware Rims 217 56 339 71
V A U G H N 97
in the context of small-scale household rituals. Again,
in contrast to earlier and later periods in the region (see
below), there is no community-wide organization of
ceremonial activities. There are no public or ceremonial
spaces at Marcaya, and it appears that some form of
ritual may have transpired at the household scale.
In summary, the community at Marcaya is charac-
terized by domestic architecture consisting of round
houses and associated patios. Each patio group is the
locus of a single archaeological household and contains
1) material correlates of food processing, preparation,
and storage; 2) artifacts relating to lithic and textile
production; and 3) a large percentage of Nasca poly-
chrome newares in the ceramic assemblage. Moreover,
each household was economically independent, as each
patio group examined featured redundant archaeologi-
cal assemblages and features.
REGIONAL COMPARISON
The utility of these ndings is clearly limited if analysis
is restricted to a single site in the region. Comparison
with other Early Nasca sites is required in order to
evaluate if these composite features are diagnostic of
the SNR. Although excavations to date have only been
conducted at a single domestic site, intensive regional
survey of the SNR has revealed that the domestic
architecture of contemporaneous sites follows a pat-
tern similar to Marcaya.
Of the fty or so Early Nasca habitation sites that
have been recorded by Schreiber in the SNR, all have
the same characteristic architecture described above
(Schreiber 1988; Schreiber and Isla Cuadrado 1996).
Specically, all are composed of agglutinated and dis-
persed structure clusters consisting of round houses
with associated patios forming patio groups. For
example, gure 10 is an oblique aerial photo taken of
an Early Nasca site located in the upper Tierras Blancas
River Valley. Again, lack of deposition facilitates inspec-
tion of domestic architecture, and the architectural
patterns revealed resemble those at Marcaya. There
are structure clusters that are composed of small round
structures attached to large ovoid structures. A large
circular structure in the southeastern corner of the site
is a notable deviation from this architectural pattern.
Though excavations are needed to verify its function, I
suspect that this structure may have been a corral.
Structures at these domestic sites are made from
unmodied to slightly modied eldstone, and, at least
from what is visible on the surface, adobe was not used
in their construction (Schreiber 1988; gure 11). The
sizes of the patio groups appear to be comparable to
those at Marcaya, and the sites themselves range in size
from one half hectare to three hectares (Schreiber and
Isla Cuadrado 1996). The condition of sites through-
out the region varies considerably, as some have poorly
preserved walls and appear to be severely deated, while
the structures of others are in excellent condition (gure
11). Collomas are visible on the surface of various Early
Nasca sites in the region as well (gure 12).
Excavations at Marcaya have demonstrated that the
high quantity of neware on the surface of sites in the
region is indicative of subsurface deposits. It is expected
that excavations in contemporaneous sites will verify
these ndings with similar artifact assemblages.
The remarkably consistent architectural forms of this
time period are completely distinct from the preceding
Early Horizon (ca. 800 BCAD 1) residential sites in
the region, which were composed of house terraces
Figure 10. Aerial photo of Early Nasca site located in the upper
Tierras Blancas River Valley. Architectural foundations visible in the
photo and the valley margin have been traced. Note the large corral-like
structure in the southeastern corner of the site. Photo is courtesy of
Katharina Schreiber.
98 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
(Schreiber and Isla Cuadrado 1996). Moreover, the
changing of architectural forms between the Early
Horizon and the Early Intermediate period is coupled
with an abrupt change in settlement pattern (Schreiber
1998; Schreiber and Lancho Rojas 1995).
One notable change between settlements of these
phases of the Early Horizon and Early Intermediate
period is in the public space apparently reserved for
ceremonial activities. In the Early Horizon, settle-
ments not only are composed of domestic terracing,
but often have large, open spaces situated within or
adjacent to the community. Currently it is unknown
what form of activity took place in these spaces, but it
is reasonable to speculate that they were reserved for
group-oriented ceremony. Public spaces are not found
at any Early Nasca residential site in the region, and
instead the domestic architecture described above and
the correlates of socioeconomic behavior at these sites
suggest that economic and ceremonial activities were
focused at the household scale.
The shifting pattern of group-oriented to house-
hold-oriented ceremony at domestic sites does not
indicate that group-oriented activities relating to ritual
fell out of favor in Early Nasca. Indeed, it has been
repeatedly argued that the ceremonial site Cahuachi
was the locus of group-oriented ritual activity for the
entire Nasca region during Early Nasca times (Silver-
man 1993b, 1994a). The data presented here support
this argument insofar as group-oriented ritual behavior
is not evident at residential sites contemporaneous with
Cahuachis apogee.
Following Early Nasca, Middle Nasca settlement
changed again as people began to inhabit the middle
valley for the rst time in prehistory (Schreiber and
Lancho Rojas 1995). Architecturally, Middle Nasca
is transitional between Early and Late Nasca in that
structures become slightly more rectilinear, but houses
with patios are still the predominant architectural
form (Schreiber and Isla Cuadrado 1996). Late Nasca
architectural forms consist of completely rectilinear
compounds agglutinated into large towns (Schreiber
1994; Schreiber and Isla Cuadrado 1996). These
changing architectural forms may have been a result
of cultural disruption after the fall of the inuence
of Cahuachi and a concomitant increase in conict
(Schreiber 1998; Schreiber and Lancho Rojas 1995).
To briey sum up the major conclusions thus far,
while house forms and community organization change
drastically between the Early Horizon and Early Inter-
mediate period, they change more from Early to Middle
to Late Nasca in the southern Nasca region. It appears
that for the Early Nasca time period at least, the archi-
tectural forms described previously are diagnostic of the
local group of the region. The pattern of round houses
Figure 11. Architecture of a typical house at an Early Nasca site located in the Tierras Blancas River Valley.
V A U G H N 99
adjoined to ovoid patios is clearly present in other
Early Nasca sites in the area. A household archaeology
approach has enabled the analysis of the archaeological
community, and it has provided a means by which com-
munity and household organization can be compared
across time periods.
DISCUSSION
An attempt has been made to demonstrate that by
employing a household archaeology approach, the
domestic architecture of Early Nasca sites in the
southern Nasca region can be used as an independent
measure of the archaeological manifestation of ethnic-
ity. The Early Intermediate period on the south coast
of Peru has often and erroneously been treated synony-
mously with the developments of Nasca society. This is
primarily the result of the presence of Nasca neware
ceramics at sites from the Pisco Valley in the north to
the Acar Valley in the south. Research has suggested,
however, that distinct local traditions existed outside
the Ica-Grande drainages (Carmichael 1992b; Massey
1986; Peters 198788; Silverman 1995:16, 1997; Valdez
1998; Wallace 1986).
While the presence of polychrome newares in the
SNR, augmented by a distinct domestic architecture,
appears to identify local groups, it is unclear how preva-
lent this pattern is throughout the Ica-Grande region
(Carmichael 1992b; Schreiber 1998; Silverman 1994a,
1997). As stated previously, it has been noted that there
is a high degree of neware present in the residential sites
of the Ica-Grande region (Browne and Baraybar 1988;
Massey 1986; Reindel and Isla 1998; Schreiber and Isla
Cuadrado 1996; Silverman 1993a). It is unknown, how-
ever, whether surface ceramics correlate with domestic
occupations, as excavations have been limited. Outside
the Ica-Grande heartland, Nasca ceramics are present
at sites, though these appear to be trade pieces and elite
goods (Carmichael 1992b; Valdez 1998).
Silverman has recorded a number of sites in the
Ingenio Valley, many of which date to the Early Inter-
mediate period (phases 15) (Silverman 1993a). The
descriptions of architecture and maps of Nasca 3 sites
in Ingenio bear some resemblance to the architectural
canons described here for Early Nasca sites in the
southern Nasca region, though there is some varia-
tion. Silverman describes Early Nasca habitation sites
recorded in Ingenio as more or less large, with almost
rectangular structures. Furthermore, these structures
are found on terraces made of eldstone that follow the
topography of the hills upon which they are situated
(1993a:112, my translation). However, planviews of the
sites show that there are some structure clusters that
look similar to the patio groups in the southern Nasca
region, and some of the architectural walls present in a
few structures are slightly curvilinear.
Figure 12. Collomas on the surface of an Early Nasca site located in the Tierras Blancas River Valley.
100 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
S. A. Massey, in her study of the upper Ica Valley,
has noted the presence of large amounts of neware in
domestic sites dating to Nasca 3 and 4 (Early Nasca in
this paper; 1986:203), but she maintains that the neware
is elite pottery (1986:201). Massey reports that the
architecture of domestic sites in the upper Ica Valley is
characterized by irregular structures made of eldstone
and sometimes painted adobe. Planviews of several of
these sites bear some resemblance to the architectural
patterns at Early Nasca sites (Massey 1986: appendix 2,
gure 4.19, site 12J08). Most sites dating to EIP 3 and 4,
however, appear to have rectilinear architecture incorpo-
rated into terrace rows, although it is difcult to decipher
the exact architectural formations in Masseys idealized
planviews (for example, Massey 1986: appendix 2, gure
4.18, site 14I01, and gure 4.20, site 11J07).
Though the architectural canon of round houses
and patios organized into patio groups appears to be
present throughout the Ica-Grande region, the above
studies show that its prevalence varies. Moreover, there
are considerable differences in settlement patterns from
valley to valley (Browne 1992; Browne and Baraybar
1988; Massey 1986; Reindel and Isla 1998; Schreiber
and Lancho Rojas 1995; Silverman 1993a), as well as
a possible variability in pottery styles (Proulx 1968;
Silverman 1995:16). Associated artifact assemblages
are not known from these regions at present, as exca-
vation data are lacking. However, following the lead
of Silverman (1995:16), regional variation should be
considered when making valley to valley comparisons,
and until careful analysis of both domestic architecture
and excavated artifact assemblages are made, the assess-
ment of an Ica-Grande regional group will necessarily
be preliminary.
Outside of the Ica-Grande region, Valdez (1998)
has analyzed the relationships between Acar and the
Nasca culture during the EIP. He nds that although
there was probably signicant interaction between the
regions, the Acar region developed as a distinct local
tradition called Huarato (Valdez 1998:149). Further-
more, the presence of Early Nasca neware at sites such
as Tambo Viejo, which previously was thought to be an
Early Nasca urban settlement (Rowe 1963), is explained
by elite interaction and trade pieces, since the dominant
style of artifact at the site is the Huarato style, which
is composed of relatively plain pottery (Valdez 1998:
98). Although the site spans the Early Horizon through
the Colonial period, and the Early Intermediate period
occupation is difcult to isolate, the architecture of this
large agglutinated settlement is completely distinct
from the architecture characteristic of Early Nasca
domestic sites described above. Indeed, the architecture
of the Huarato tradition, which is contemporaneous
with Early Nasca, consists mainly of rectilinear struc-
tures usually surrounded by large walls (Valdez 1998:
150). Thus, it appears that in the Acar region a distinct
ethnic group, evident by both domestic architecture and
associated artifact assemblages, is distinguishable from
the ethnic group that is dened in the SNR.
CONCLUSIONS
Beyond providing a simple measure of identifying
the local group, a household archaeology approach
facilitates the evaluation of economic and political
relationships within archaeological settlements. What
can be said about the economic and political relations
manifest in Early Nasca settlements, and how do these
compare to settlements of earlier and later time periods
and from other regions? The salient economic pattern
at Early Nasca sites in the southern Nasca region is
that households are economically independent. Every
patio group excavated at Marcaya contained the req-
uisite components to sustain domestic units. Each had
evidence for food processing (grinding stones, ollas),
subsistence-goods storage (collomas, jars), and lithic
and textile production. All households were involved
in some form of ritual activity, as a large quantity of
Nasca neware is present in each patio group. In con-
trast, there is no evidence for socioeconomic or ritual
activities that are organized at the community level.
Community activities in both Early Horizon and
later Early Intermediate period domestic sites in the
region, however, are present in the form of special
architectural features and large public spaces. This
development may be the result of spatially and tempo-
rally distinct socioeconomic and ritual activities. Until
further excavations are conducted at these sites, it would
be premature to speculate on the nature of these com-
munity activities; employing similar methodological
approaches as outlined in this paper would facilitate an
evaluation of these sites.
To conclude, although the domestic architecture and
associated artifact assemblages indicate that there is a
denable ethnic group in the southern Nasca region
during the Early Nasca period, at this stage of research
it is unclear how representative of the entire Ica-Grande
region this pattern is. Previous research has shown
unmistakably that outside the Ica-Grande region, dis-
tinct local traditions thrived with only occasional contact
V A U G H N 101
with Nasca. The primary goal of the settlement survey
projects conducted within the region was to explicate
more fully the nature of Nasca society (see Silverman
1995 for a summary). While there are notable intravalley
differences in settlement patterns and pottery styles, it
is urged that additional excavations in other regions are
conducted employing the approach taken here so that
these differences can be further elucidated.
ACKNOWLEDGEMENTS
Fieldwork at Marcaya was funded by a Fulbright
Hays Doctoral Dissertation Research Abroad Fellow-
ship #P022A70041 and a Wenner Gren Foundation
for Anthropological Research Small Grant #6227. I
am grateful to the Instituto Nacional de Cultura of
Peru, which kindly granted permission to conduct
eldwork at Marcaya (permit #2063-97). Schreiber,
van Gijseghem, Reycraft, Vaughn, and an anonymous
reviewer read earlier versions of this paper, and the nal
draft has beneted greatly from their comments. This
paper has also beneted from previous discussions that
I had with Barbara Wolff. All inaccuracies, of course,
are my own. I would also like to thank Reycraft for the
invitation to write a paper for this volume.
REFERENCES
Aldenderfer, M., ed.
1993 Domestic architecture, ethnicity, and complementarity in
the south-central Andes. Iowa City: University of Iowa
Press.
Aldenderfer, M., and C. Stanish
1993 Domestic architecture, household archaeology,
and the past in the south-central Andes. In Domes-
tic architecture, ethnicity, and complementarity in the
south-central Andes, edited by M. Aldenderfer, 112.
Iowa City: University of Iowa Press.
Barth, F.
1969 Ethnic groups and boundaries: The social organiza-
tion of culture difference. Boston: Little, Brown and
Company.
Bawden, G.
1982 Community organization reected by the house-
hold: A study of pre-Columbian social dynamics.
Journal of Field Archaeology 9 (2): 165181.
Bender, D.
1967 A renement of the concept of household: Fami-
lies, co-residence, and domestic functions. American
Anthropologist 69:493504.
Bermann, M.
1994 Lukurmata: Household archaeology in prehispanic
Bolivia. Princeton, NJ: Princeton University Press.
1997 Domestic life and vertical integration in the Tiwanaku
heartland. Latin American Antiquity 8 (2):93112.
Bermann, M., and J. E. Castillo
1995 Domestic artifact assemblages and ritual activities in
the Bolivian Formative. Journal of Field Archaeology
22 (4): 389398.
Browne, D. M.
1992 Further archaeological reconnaissance in the Prov-
ince of Palpa, Department of Ica, Peru. In Ancient
America: Contributions to New World archaeology,
edited by N. J. Saunders. Oxbow Monograph 24.
Oxford: Oxbow Books.
Browne, D. M., and J. P. Baraybar
1988 An archaeological reconnaissance in the Province of
Palpa, Department of Ica, Peru. In Recent studies in
pre-Columbian archaeology, edited by N. J. Saunders
and O. de Montmollin, 299325. BAR International
Series 421. Oxford: British Archaeological Reports.
Carmichael, P. H.
1988 Nasca mortuary customs: Death and ancient soci-
ety on the south coast of Peru. Ph.D. dissertation,
University of Calgary.
1992a Interpreting Nasca iconography. In Ancient images,
ancient thought: The archaeology of ideology, proceed-
ings of the 23rd annual Chacmool Conference, edited
by A. S. Goldsmith, S. Garvie, D. Selin, and J. Smit,
187197. Calgary: Department of Archaeology,
University of Calgary.
1992b Local traditions on the south coast of Peru during
the Early Intermediate period. Willay 37/38:46.
1994 The life from death continuum in Nasca imagery.
Andean Past 4:8190.
1995 Nasca burial patterns: Social structure and mortuary
ideology. In Tombs for the living: Andean mortuary
practices, edited by T. Dillehay, 161189. Washing-
ton, D.C.: Dumbarton Oaks Research Library and
Collection.
1998 Nasca ceramics: Production and social context.
In Andean ceramics: Technology, organization, and
approaches, edited by I. Shimada, 213231. Phila-
delphia: University of Pennsylvania Museum of
Archaeology and Anthropology.
Clarkson, P.
1990 The archaeology of the Nazca Pampa: Environmen-
tal and cultural parameters. In The lines of Nazca,
edited by A. Aveni, 117172. Philadelphia: The
American Philosophical Society.
102 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Conlee, C.
2000 Late prehispanic occupation of Pajonal Alto, Nasca,
Peru: Implications for imperial collapse and soci-
etal reformation. Ph.D. dissertation. Department
of Anthropology, University of California, Santa
Barbara. University Microlms, Ann Arbor.
Conlee, C., and K. J. Vaughn
1999 The development of spinning technology and imple-
ments in the Nasca region. Paper presented at the
Institute of Andean Studies 39th annual meeting,
Berkeley, California.
DeLeonardis, L.
2000 The body context: Interpreting early Nasca decapi tated
burials. Latin American Antiquity 11 (4): 363386.
Emberling, G.
1997 Ethnicity in complex societies: Archaeological
perspectives. Journal of Archaeological Research 5 (4):
295344.
Gayton, A. H., and A. L. Kroeber
1927 The Uhle pottery collections from Nazca. University of
California Publications in American Archaeology
and Ethnology 24 (1): 146. Berkeley: University
of California Press.
Goldstein, P.
1993 House, community, and state in the earliest Tiwan-
aku colony: Domestic patterns and state integration
at Omo M12, Moquegua. In Domestic architecture,
ethnicity, and complementarity in the south-central
Andes, edited by M. Aldenderfer, 2541. Iowa City:
University of Iowa Press.
Jones, S.
1997 The archaeology of ethnicity: Constructing identities in
the past and present. New York: Routledge Press.
Kroeber, A. L.
1944 Peruvian archeology in 1942. Viking Fund Publication
in Anthropology 4. New York: Viking Fund.
Kroeber, A. L., and D. Collier
1998 The archaeology and pottery of Nazca, Peru: Alfred
L. Kroebers 1926 expedition. Edited by P. H.
Carmichael. Walnut Creek, CA: AltaMira Press.
Massey, S. A.
1986 Sociopolitical change in the upper Ica Valley, 400
BC to AD 400: Regional states on the south coast of
Peru. Ph.D. dissertation, Department of Anthropol-
ogy, University of California, Los Angeles.
Murra, J. V.
1972 El control vertical de un mximo de pisos ecolgi-
cos en la economa de las sociedades Andinas. In
Visita de la provincia de Lon de Hunuco en 1562,
427476. Documentos por la Historia y Etnologa
de Hunuco y la Selva Central 2. Hunuco: Univer-
sidad Nacional Hermilio V.
1985 El archipelago vertical revisited. In Andean ecology
and civilization : An interdisciplinary perspective on Andean
ecological complementarity, papers from Wenner-Gren
Foundation for Anthropological Research Symposium 91,
edited by Y. Masuda, I. Shimada and C. Morris, 313.
Tokyo: University of Tokyo Press.
ONERN
1971 Inventario, evaluacin, y uso racional de los recursos
naturales de la costa: Cuenca del Ro Grande (Nazca).
Lima: Ocina Nacional de Evaluacin de Recursos
Naturales.
Oreci, G.
1993 Nasca: Arte e societa del popolo dei geogli. Milan: Jaca
Books.
Peters, A.
198788 Chongos: Sitio Paracas en el valle de Pisco. Gaceta
Arqueolgica Andina 16:3034.
Proulx, D.
1968 Local differences and time differences in Nasca pottery.
University of California Publications in Anthropol-
ogy 5. Berkeley: University of California Press.
1970 Nasca gravelots in the Uhle collection from the Ica Valley,
Peru. Research Reports 5. Amherst: Department of
Anthropology, University of Massachusetts.
1983 The Nasca style. In Art of the Andes: Pre-Colum-
bian sculptured and painted ceramics from the Arthur
M. Sackler collections, edited by L. Katz, 87105.
Washington, D.C.: Arthur M. Sackler Foundation
and the AMS Foundation for the Arts, Sciences and
Humanities.
Reindel, M., and J. Isla
1998 Informe nal: Proyecto Palpa 1998. Instituto Nacio-
nal de Cultura de Peru, Lima.
Reycraft, R. M.
1998 The Terminal Chiribaya project: The archaeology of
human response to natural disaster in south coastal
Peru. Ph.D. dissertation, Department of Anthro-
pology, University of New Mexico, Albuquerque.
University Microlms, Ann Arbor.
Rice, D. S.
1993 Late Intermediate period domestic architecture
and residential organization at La Yaral. In Domes-
tic architecture, ethnicity, and complementarity in the
south-central Andes, edited by M. Aldenderfer, 6682.
Iowa City: University of Iowa Press.
Rowe, J. H.
1963 Urban settlements in ancient Peru. awpa Pacha 1:
127.
V A U G H N 103
Schreiber, K. J.
1988 On revisiting Huaca del Loro: A cautionary note.
Andean Past 2:6979.
1990 Proyecto Nasca Sur 1990: Informe nal. Report sub-
mitted to the Instituto Nacional de Cultura de Peru.
1994 Proyecto Nasca Sur 1994: Informe nal. Report sub-
mitted to the Instituto Nacional de Cultura de Peru.
1998 Afterword. In The archaeology and pottery of Nazca,
Peru: Alfred L. Kroebers 1926 expedition, edited by
P. H. Carmichael, 261270. Walnut Creek, CA:
AltaMira Press.
1999 Regional approaches to the study of prehistoric
empires: Examples from Ayacucho and Nasca, Peru.
In Settlement pattern studies in the Americas: Fifty years
since Vir, edited by B. R. Billman and G. M. Fein-
man, 160171. Washington, D.C.: Smithsonian
Institute Press.
Schreiber, K. J., and J. Isla Cuadrado
1996 Proyecto Nasca Sur 1996: Informe nal. Report
submitted to the Instituto Nacional de Cultura de
Peru, Lima.
Schreiber, K. J., and J. Lancho Rojas
1995 The puquios of Nasca. Latin American Antiquity 6
(3): 229254.
Silverman, H.
1993a Patrones de asentamiento en el valle de Ingenio,
cuenca del Ro Grande de Nasca: Una propuesta pre-
liminar. Gaceta Arqueolgica Andina 7 (23): 103124.
1993b Cahuachi in the ancient Nasca world. Iowa City: Uni-
versity of Iowa Press.
1994a The archaeological identification of an ancient
Peruvian pilgrimage center. World Archaeology 26
(1): 118.
1994b Paracas in Nazca: New data on the Early Horizon
occupation of the Rio Grande de Nazca drainage.
Latin American Antiquity 5:359382.
1995 Recent archaeological investigations on the south
coast of Peru: Critique and prospects. Journal of the
Steward Anthropological Society 23(12): 1341.
1996 The Formative period on the south coast of Peru: A criti-
cal review. Journal of World Prehistory 10 (2): 95147.
1997 The first field season of excavations at the Alto
del Molino site, Pisco Valley, Peru. Journal of Field
Archaeology 24 (4): 441458.
Stanish, C.
1989 Household archaeology: Testing models of zonal
complementarity in the south-central Andes.
American Anthropologist 91:724.
1992 Ancient Andean political economy. Austin: University
of Texas Press.
Stanish, C., E. de la Vega, and K. L. Frye
1993 Domestic architecture on Lupaqa area sites in
the Department of Puno. In Domestic architecture,
ethnicity, and complementarity in the south-central
Andes, edited by M. Aldenderfer, 8393. Iowa City:
University of Iowa Press.
Uhle, M.
1914 The Nazca pottery of ancient Peru. Proceedings of the
Davenport Academy of Sciences 13:116.
Valdez, L.
1998 The Nasca and the Valley of Acar: Cultural interac-
tion on the Peruvian south coast. Ph.D. dissertation,
University of Calgary.
van Gijseghem, H.
2001 Household and family at Moche, Peru: An analysis of
building and residence patterns in a prehispanic urban
center. Latin American Antiquity 12 (3): 257273.
Vaughn, K. J.
1999a Early Intermediate period complexity on the south
coast of Peru: New perspectives from Marcaya.
Paper presented at the Society for American Archae-
ology, 64th annual meeting, Chicago, Illinois.
1999b Early Intermediate period households on the south
coast of Peru: Excavations at Marcaya, an early Nasca
domestic site. Paper presented at the Institute of Andean
Studies, 39th annual meeting, Berkeley, California.
2000 Archaeological investigations at Marcaya: A village
approach to Nasca sociopolitical and economic
organization. Ph.D. dissertation, Department of
Anthropology, University of California, Santa Bar-
bara. University Microlms, Ann Arbor.
Vaughn, K. J., and M. Glascock
i.p. Exchange of Quispisisa obsidian in the Nasca region:
New evidence from Marcaya. Andean Past 7.
Vaughn, K. J., and H. Neff
2000 Moving beyond iconography: Neutron activation
analysis of a ceramic sample from Marcaya, an early
Nasca domestic site. Journal of Field Archaeology 27
(1): 7590.
Wallace, D. T.
1971 Sitios arqueolgicos del Per (segunda entrega): Valles
de Chincha y de Pisco. Arqueolgicas 13:4131.
1986 The Topar tradition: An overview. In Perspectives on
Andean prehistory and protohistory, edited by D. P. Kvi-
etok and D. H. Sandweiss, 3548. Ithaca, NY: Cornell
University Latin American Studies Program.
Wells, P.
1998 Review of the archaeology of ethnicity: Construct-
ing identities in the past and the present by Sin
Jones. American Antiquity 63 (2): 347348.
104
Without a doubt, the art and manufacture of textiles
constitute one of the most signicant achievements of
the ancient peoples of Peru.
(Lumbreras 1974: 17).
W
hile it is well known that contemporary Peruvian
weavers are descendants of a long and extensive
textile heritage, the problem arises as to whether con-
temporary textiles continue to communicate ethnicity
for their makers and wearers. In Paracas, as in other
pre-conquest Peruvian cultures, complex ideas were
transmitted in the cloth of woven garments (Paul 1990:
16). Paul concludes that dress is a form of communi-
cation, functioning as an overt expression of identity:
anyone who is literate in the styles of weaving, in the
codes of dress, and in the designs woven into cloth
in these communities is able to read instantaneously
about the wearer. Embedded in terms of apparel are
visible indicators of such things as ethnic group, ayllu
afliation, cosmology, economic status, gender, age,
family ties, marital status, and ofces held within the
community (1990:17). This paper will investigate the
continued use and making of hand-woven contem-
porary textiles as expressions of identity and ethnicity
for the Quechua-speaking people who live in the rural
landscape surrounding the sacred peak referred to as
Apu Ausangate (6,392 meters), in the Cordillera Vilca-
nota range of mountains southeast of Cuzco, Peru.
Farming potatoes, tending large herds of alpacas
and llamas, and back-strap loom weaving are the main
activities sustaining local inhabitants. Their language is
Quechua with rudimentary Spanish as a second language
spoken mostly by children who have attended schools and
men who travel for work or commerce. With the intro-
duction of rural schools that aspire to integrate campesinos
into the national economy and market sales agents who
sell synthetic materials and clothing, along with aniline
dyes, a paradoxical question arises. Do weavers still
possess the technical skills and ancestral knowledge
necessary to execute the intricate design vocabulary to
express ethnicity commonly identied as being from the
region of Ausangate and recognizable as a distinctive
style of clothing assembled from a variety of locally made
CHAP T E R 7
Cultural Communication of
Ethnicity through Clothing:
T HE QOCHA- L AKE S Y MBOL I N
CONT E MP ORARY T E XT I L E S F ROM
AUS ANGAT E , P E RU
ANDREA M. HECKMAN
H E C K M A N 105
textiles? Girls and women are the weavers who craft the
textiles with an understood status system based on their
expertise and knowledge of local symbols.
This projects primary goal was to document the
motifs and visual metaphors used to express complex
cultural beliefs through symbols, and the structures
of cloth related to their contextual daily use and
ritual purposes. In rural areas near Cuzco, Peru, some
communities, like that in the Ausangate region, have
distinctive clothing identifying them generally as Que-
chuas. While many ethnographers and linguists reserve
the word Quechua only for the spoken language, people
like Eloy Turpo in the Ausangate area say, we are Que-
chuas (1996:personal conversations). Many academics
generally use the term campesino (Spanish for country
person) for rural inhabitants; however here it is not the
appropriate term. The preferred term in Ausangate is
Quechua, which includes the linguistic distinction of
speaking the same native language but also commu-
nicates a sense of identity and pride associated with a
shared ancestry. Quechuas here also refer to themselves
as runakuna (Q. for mankind) who speak runasimi (Q.
for the Quechua language). As ethnographers we have a
disciplinary obligation to use the internal categories of
identity revealed through the methodology we employ:
the art of eldwork. Lively discussions about ancestral
designs and their Quechua names were encouraged
during informal gatherings and interviews by the pre-
sentation of a sample set of locally woven moneybags.
Shoulder cloths (kaypinas), belts (chumbi), bags (chuspas),
ponchos, and other variations of textiles display a range of
similar motifs, as they are used for daily wear, rituals, pil-
grimages, and festivals. Their active use communicates a
sense of pride and ethnicity associated with the retention
of cultural traits identied as Quechua. While culture is
dynamic, changes are incorporated into a bedrock of tra-
ditional ideas related to Andean worldview. Today, roads
introduce manufactured market goods from Cuzco, and
electricity is followed by televisions, but the often reiter-
ated view that the intrusion of the modern western world
with its commercial values and commodities signals the
end of traditional Andean culture is simplistic and largely
inaccurate (Bawden 2003).
WRITTEN LANGUAGE AND VISUAL
REPRESENTATION
Within the quipu, we can show by the groupings of
these categories what the priorities were in the think-
ing of Andean peoples. First come the people, second
the camelids, then the cloth, fourth the ceramics and
so on. For a people who have made a natural refrigera-
tor, survive at freezing temperatures, build bridges as
a community task and still weave and maintain a status
system based on the neness of cloth, to investigate this
world, we must follow Andean thought and language
patterns. You have a logic, not our categories nor a
historical one but Andean logic.
(Murra:1980)
While we can argue a continuance of communication
of worldview and ethnicity through the consistent use
of textiles, both in design and structural textile infor-
mation, this is not to claim that the design vocabulary
of today is consistent with symbolic meaning from the
past. While Ausangate weavers may say it is the pallay
(designs) of our ancestors, it is difcult to make de-
nite correlations of meaning to similar motifs evident
in Inka cloth because of the lack of a written history
with detailed notations about textiles. Rather than ask
why a written language was never invented, the more
provocative question, in light of the many accomplish-
ments of early Peruvian civilizations, is what did they
use instead of writing to communicate worldview and
reinforce ethnicity?
The arid coastal desert stretching the length of
Peru between the Pacic Ocean and the mountains
has preserved a rich cache of material evidence.
Coastal bundles found in the Paracas peninsula display
advanced textile techniques such as sprang, embroidery,
and feather work, which evidenced a highly sophisti-
cated knowledge of textile and fabric expertise (Bird
and Bird 1949). Wari-Tiwanaku symbols placed in
cloth demonstrate intentional repetitions of units with
slight variations that displayed interesting contractions
and expansions of related units appearing to represent
elaborate metaphors (Bird and Bird 1949; Sawyer 1963).
Given the extensive labor, detail, and skill of the crafts-
manship, could these precise contractions, expansions,
and juxtaposition of individual units have metaphori-
cally communicated changes in historical or political
events or epochs such as the rise and fall of administra-
tive power structures (Heckman 2003; Stone-Miller
1992)? The exact meanings placed in ancient cloth by
these skillful artisans remain an enigma.
The high mountainous Andean regions, with annual
rainy seasons, did not afford the same natural preserva-
tion for cloth; thus the range of textile remains there
is limited. Andean logic and local knowledge inform
us that cloth performs many important roles besides
bodily warmth and protection in rugged mountain
106 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
environments and cold climates. During the Inka
Empire, cloth symbolized community and state wealth,
personal prestige or status, and was often used as ritual
offerings for ceremonies. Inka textile specialists such
as the chumbicamayoqs wove ne cloth designated as
honorable royal gifts given by the Inka (Murra 1962).
Textiles celebrated and accompanied major life transi-
tionsbirth, rst hair cutting, puberty rites, marriage,
yearly agricultural offerings, and deathas they still
do today in remote Quechua communities. The Inka
administration systematically organized people into
work units and maintained the careful storage of all
excess goods in state warehouses, accounting for their
tallies on quipus. While quipus demonstrate a tactile
information storage system, interestingly Inka textiles
used geometric symbolic units in relationships of duality
and inversions, often woven in black-and-white check-
erboard patterns or what appear to be signicant color
arrangements. Inka architecture, road building, and
palace and temple structures display a sense of Inka aes-
thetics related to line, form, and intentional placement
of units. Could they have been re-creating a sense of
the order of the universe in their artistic textile motifs,
the nest of which accompanied every important event?
Spanish chroniclers did record the importance of Inka
textiles, but their European sense of written history did
not allow them to understand or bother to notate the
meanings of Inka visual representation.
The quipu used a tactile arrangement of knots and
cords using a mathematical base-10 system (Asher &
Asher 1981). Using a sample of quipus from various
museums in the 1920s, Leland Locke discovered
a consistent code based on the kind of knot used
that helped him discover that knots could represent
numbers in the hundreds of thousands. A highly
trained group of Inka specialists, the quipucamayocs,
were responsible for creating color-coded cords with
carefully constructed, strategically placed knots for
decoding information and for retelling oral narratives
using quipus as mnemonic devices (Cummings 1994:
194). The detailed analysis of quipus illustrates how
Inka specialists used a tactile device to enhance oral
tradition and numerical codication.
Tocapu refers to abstract geometric designs with stan-
dardized square or rectangular units found on certain
nely woven mens tunics (unku) and womens dresses
(anaku) and cloaks worn at important Inka ceremonial
occasions. Variations of tocapu existed, including several
rows of pattern on a solid background, an upside-down
stepped pattern on top, and totally patterned garments,
sometimes even including alternating units of black-
and-white checkerboard squares or the distinctive Inka
key motif. Tocapu can be as simple as repetitious motifs
with only alternating colors or as complex as detailed
patterns totally covering the surface of a mans tunic,
back and front. Cummings (1994) stated that tocapu
probably demarcated ethnic, political, and religious
status but the Spanish chroniclers did not relate to this
abstract visual form of images as recording Inka histo-
ries. Sarmiento and Molina suggested that certain Inka
painted boards recorded tocapu design elements and
were used by Inka judges. Cummings (1994:198) said
the abstractness of the tocapus did not correspond to
the perceived ideas of the Spanish chroniclers of how an
image should look. They were not able to understand the
design information and they did not record its details.
Possibly Inka nobility did not share an explanation of
tocapu designs with the Spanish invaders. Tocapu was
dened as a royal privilege, and grants of it were highly
prized for rulers and the military (Murra 1962:64).
Eighty years after the arrival of the Spaniards, Felipe
Guaman Poma de Ayala wrote a 1,179-page letter to
the king of Spain, including 397 drawings in the letter.
His drawings are the best visual record of these textile
designs outside the actual tocapu weavings themselves.
He recorded that the tocapu consisted of rows of rect-
angular or square geometric designs using a pattern of
repetition. Inka tocapu demonstrates concepts of rep-
etition, inversions, and reversals, with the juxtaposing
of many of the unitsexpressions of duality between
black and white, colors in balanced oppositionwith
accents of red, the color generally associated with the
Sapa Inka. Rowe (1979) detected regular, precise, appar-
ent standardization in the patterns given to intentional
irregularities and an asymmetrical balance in colors,
Figure 1. Inka mens tunic-style clothing with woven tocapu patterns
as depicted by Guaman Poma.
Figure 2. A quipu being used by the chief treasurer and accountant
(Guaman Poma de Ayala).
H E C K M A N 107
numbers, and forms in the arrangement of the tocapu.
Textile specialists continue their quest for a Rosetta
stone leading to the decoding of Inka information in
woven tocapu, but few cloth examples of tocapu survive
today. Currently, MacArthur fellow and anthropologist
Gary Urton investigates Andean logic and quipus from
a mathematical approach, but we do not yet understand
the code of the tocapu. Tocapu was also painted on the
wooden drinking cups called keros, and researchers at
the University of San Antonio de Abad in Cuzco con-
tinue to work with their museums collection of over
200 wooden keros with tocapu.
Andean record keeping has taken many diverse forms
over time, from images burned and etched on gourds
depicting courtship, marriage, medicinal practices,
weaving and the arts, herding and cultivation, death,
and other rites of passage, to Moche designs painted
on ceramics showing virtually every aspect of daily and
ritual life (Donnan 1978). Material evidence lends sup-
port to the idea that past and present weavers encode
their cultural beliefs and identity into complex systems
of visual metaphors capable of communicating their
perception of the world.
THE CONTEXT OF LIFE IN AUSANGATE, PERU
Studies of indigenous textiles must necessarily be
conducted as cultural investigations that are set in a
deep contextual framework of the weavers commu-
nity, language, social organization, rituals, myths and
cosmology.
(Ortiz 1994)
Today in the Ausangate region contemporary Quech-
uas rely more on visual means of communication than
on written history books or textbooks to educate their
youth about how to exist at elevations of 3,500 meters;
in fact few such books written from the Quechua view-
point exist. Limited supplies of textbooks are primarily
written in Spanish, the second language of the region.
Transference of knowledge and cosmology through
verbal and visual means other than writing is a logi-
cal and efcient system for teaching children and the
community at large. In such an isolated area, Quechuas
have managed to make active choices and incorporate
what is efcient and discard what is not. Globalized
economy, foreign researchers, rural schoolteachers, and
engineers do not necessarily have the best solutions for
the problems facing rural Quechuas in an environment
in which they have survived for centuries.
Ausangate is about 130 kilometers southeast of
Cuzco. Human beings live at elevations between four
and ve thousand meters by farming, herding, and
weaving. Their sod, stone, and mud houses are pri-
marily for food storage and nighttime shelter, while
the majority of days are spent in activities outside in
the natural environment. They share a symbiotic rela-
tionship with their animals, which are the equivalent
of their bank account. The area is sparsely inhabited,
and those that live there are bonded by a cosmology
based on ancestral myths, rituals, and a respect for Apu
Ausangate, a great mountain spirit considered sacred to
the Inka, as chronicled by Spanish scribes (Gow 1976).
The name Ausangate refers ofcially only to the highest
peak in the Cordillera Vilcanota but is generally used
to refer to the entire region surrounding the moun-
tain. The west side of Ausangate was a large hacienda,
known as Lauramarka, owned for several hundred
years by absentee landlords living in Lima and Cuzco,
until the nineteen-seventies, when the agrarian reform
redistributed large land holdings. General reference
has often been made to the north side of Ausangate
as Lauramarka or Ocongate rather than the proper
community names such as Pacchanta, Upis, and Tinki.
Ocongate is the municipal center for the northwest side
of Ausangate, with a Sunday food and consumer market,
national police road checkpoints, judicial ofces, nota-
ries for legal documents, and a central Catholic church.
Pitumarka (Canchis) is the provincial township on the
southwest side of Ausangate, which provides legal ser-
vices and ofcial documents. It is the trailhead for access
to the south side of Ausangate via Chilca.
Figure 3. Map of Ausangate region (Heckman 2003, prepared by
Deborah Reade).
108 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
ETHNOAESTHETICS AND THE MEANINGS OF
AUSANGATE TEXTILE DESIGNS
In 1996 while doing eldwork and living in the village of
Pacchanta, on the north side of Ausangate (a four-hour,
seven-hundred-meter uphill hike from Ocongate), I had
the opportunity to participate in daily weaving activities.
After initially learning textile pallay from my primary
teacher, Maria Merma Gonzalo, and her extended fam-
ily in addition to other weavers in Pacchanta, I began
to circulate with the sample set of twenty moneybags
known as pachaqchaki. The moneybags averaged three to
ve inches in width and nine to fteen inches in length.
These types of textiles were important when the cur-
rency of Peru consisted primarily of large coins, because
Quechua clothing did not have pockets, and so the bags
were used to guard money. The Pacchanta moneybags
are distinctive, and this type of bag is now rare in other
communities surrounding Ausangate.
The use of the sample bags served several method-
ological purposes: 1. they were easily transportable; 2.
they were capsules of design information that could
be shown all at one time and arranged in an order of
merit according to local aesthetics; 3. they stimulated
the naming of specic design elements, which led to
discussions of meaning of symbols; 4. they represented
a wide variety of Pacchantan designs; and 5. they sup-
plied comparative data that could be quantied. The
sample was collected from 1990 to 1996; thus it con-
tained old examples and some unusual pieces such as the
bag made from a Cotobambas belt from the Apurimac
area nished properly like a Pacchantan moneybag.
Another moneybag was begun as a hat, then the male
knitter haphazardly turned it sideways and nished it as
a moneybag. Aesthetically this was totally unacceptable.
Maria called it trash, other weavers were disgusted
by it, and it rated lowest on the aesthetic scale. Dur-
ing the discussions of the moneybags, my presence
was minimalized. This approach lessened the reliance
on interviewing for collecting data and the associated
problems of answers as reaction to how a question is
posed during the interview.
The primary motif in almost all Pacchanta weavings
was a diamond-shaped pattern called a qocha, represent-
ing one or more lakes. It became critically apparent that
geography and worldview inuenced what the weavers
chose from their design repertoire. They were judged
by their knowledge of aesthetics and skill. They knew
the mythology relating to symbols such as the lake pat-
tern, which represented various lakes fed directly from
the sacred peak or underground springs. The variety
of colors related to mineral content, shapes, series,
and types of lakes were being created in the textiles.
A specic terminology was used not only to describe
Figure 4. Two of the twenty moneybags composing a sample set of
local weaving designs.
Figure 5. Maria Merma Gonzalo, from Pacchanta, Ausangate,
embroiders various colored yarns on top of the huasca qocha pallay to
embellish her sons poncho that she wove for him. She wears culturally
identiable clothing from her village and the NW side of Ausangate.
H E C K M A N 109
the variations in highland lakes but also to embed an
understanding of the importance of water and the lakes
as repositories of Ausangates sacred waters. Weavers
embellished the lakes with sequins to signify sunlight
shimmering on the lakes surfaces. They expressed
composite ideas by placing one design element inside
another, representing ideas such as a potato plant nur-
tured by the water from the lake. Potatoes and tubers
are the only crops grown in Pacchanta. The weavers
composed complex symbols and multilayered ideas
related to worldview and mythology.
Technically speaking, weavers are constructing
warp-faced weaves, that is the warps are much more
closely spaced than the wefts so that the weft does not
show on either surface of the fabric at all but passes
between the two layers of warps which interchange
to form the woven structure (Rowe 1977:14). This
process includes picking particular supplemental warps
to create images made by oating surface yarns, or by
hiding them underneath. Women are the ne weavers
and bearers of intricate design knowledge for expres-
sion of ethnicity. Men are knitters and they weave
simple weft-faced saddle blankets or coarse fabrics
for skirts and pants, mens plain carrying cloths and
formerly mens shirts along with ropes and slings. Men
could articulate design names, but if disagreements
occurred, they were quickly resolved by consulting
the older women.
Young girls begin weaving by the ages of eight to
ten years old. They rst learn plain warp-face weave,
graduating from learning one design to another in a
specied order from their aunts, mothers, and grand-
mothers. They gain social status from the increased
number of pallay designs they learn and by how nely
made they are. All family members engage in spinning
on a wooden distaff about one foot in length known as
a drop spindle or puska or phuska (from the Quechua
verb puskay or phuskay), a time-consuming activity that
can be efciently performed while tending animals or
walking up to work in the elds. Weavers have no
sketches, relying on their memories to conceptual-
ize designs; and they are capable of conversing while
silently counting the mathematical relationships of
picks for the difcult designs.
During these conversations, weavers confirmed
that lakes called qochas were places where water was
stored or contained, enclosed and protected. Specic
vocabulary for lake patterns was identied as follows,
Figure 6. Detail of a womans carrying cloth from Pacchanta near Hacienda Lauramarka and Ocongate, NW side of Ausangate. This
textile displays multiple lake (qocha) diamond designs with owers, possibly potato owers positioned inside the lakes, which form
enclosures for them.
110 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
with variations for number, colors, and sizes. Variations
in spellings occurred due to the spoken language being
converted into writing and the general disagreements
over spellings amongst linguists.
Huasca qocha: singular wide lakes
Rumpu qocha: singular long vertical lakes
Sarta qocha: a series of repetitive lakes
Rurun qocha: a variation somewhat like huasca
qocha
Sinku qocha: more shimmering lake
Weisto qocha: half lakes in a series of vertical
complementary patterns
Huacra qocha: large half diamonds in a series
Hatun qocha: extremely large lake
Weavers identied other simple design motifs like
tika (flowers), roses, and inti (sun) inside the lake
design. When combined with other elements, the
vocabulary conveyed cosmological information beyond
simple icons related to worldview. The potato ower
inside the lake waters conveyed the hope for good crops
(CCarita 1996:personal conversations). The borders of
textiles were often described as having chaki (feet) and
nawi (eyes). Qenko (zigzag) was mentioned as protection
for the person wearing or using the textiles, according
to local lore (for more details, see Heckman 1997). Men
knitted designs into chullo earap hats that included
alqo yupi, dogs paws; llamas footprints; bandas, diago-
nal half diamonds; listas, lines of solid colors; viscachas,
an Andean rodent; trigo, wheat; pampa, large open
solid colored areas; and of course many lake patterns.
The qocha design was important in areas as distant
as Carabaya and Marcusani. In Paucartambo, Qero,
Lares, and CCachin, as in Ausangate, the primary motif
on the front of mens chullos (a type of knitted hat) is a
hatun, or large qocha pallay.
An unusual body metaphor was represented in Pac-
chantan weavings. In textiles woven in two pieces, such
as carrying cloths (kaypina) and ritual cloth (unkuna)
for offerings to Apu Ausangate for health and well
being, the seam where the two sides are joined is
called a sonqo, or heart. Several weavers elaborated that
when the two parts are stitched together, they form a
complete physical body. The nawi on the sides of the
textiles are believed to protect the wearer through
sight. The chaki were placed on the four corners of
the cloth for it to be properly nished. Oddly, body
parts are arranged unusually, in some metaphoric
relation to the human body, but weavers insinuated
that this was appropriate. These and other stories
were intriguing, but the most valuable information in
talking with Pacchantan families was how the pallay is
worn actively in daily use and rituals to help reinforce
local myth and worldview as conrmation of ones
identity and ancestral knowledge.
THE MYTH OF THE WATER CYCLE REINFORCED
IN CLOTH
The natural world is interrelated, and it is known that
water is fundamental to the continuation of life. Myth
explains the cycle of water to the ocean and the annual
return as rainfall on the mountains that nourishes the
crops in the rainy season. Lakes are natural storage
facilities that catch and store this water, either from
rain or snow melt-off, or natural springs that provide
water for animals and humans during the annual dry
season. Lake patterns in textiles are indicators of this
fundamental concept of the cyclical nature of water in
Andean worldview.
The word qocha is multivocal, also referring to
unusual bowl-shaped ceramics used in Inka and colonial
Peru for ritual drinking of chicha, the local corn beer, an
important reciprocal act. Flores-Ochoa explains that the
qonopa or illa, usually in the shape of a llama or alpaca, has
a round opening on the back for offerings of animal fat;
in the illa this hole is called a qocha (1968:279). Moseley
discusses the qocha as a man-made reservoir for agricul-
tural water storage as follows:
Around the margins of Lake Titicaca soil was mounded
up to create ridged elds. Long planting ridges were
separated by deep furrows that held standing water.
Further from the lake, people dug great pond-like
structures called cochas. After lling with wet season
runoff, their sides and bottoms were farmed as their
water levels lowered during the dry season (1992:30).
In Pukara . . . substantial investment in agriculture
included ridged elds and cochas. Rainfall was seasonal
and uncertain and had to be assisted by the construction
of enormous, shallow ponds called cochas (1992:150).
Qochas are absolutely essential to agricultural suc-
cess when farmers, such as in Pacchanta, have only one
growing cycle per year. When Quechuas in Ausangate
weave qocha pallay, they are wearing their cultural
knowledge by identifying with past beliefs about the
important role of lakes and the cyclical nature of water.
The word qocha preceded by mama- forms the com-
posite word mamaqocha, meaning the sea (the mother
ocean) and Lake Titicaca, the source of the Inka origin
myth. Water is believed to originate with the Mother
H E C K M A N 111
Lake, the ocean. It then cycles up into the mountains
and is brought to the people by the rivers or under-
ground streams and then returns to the source, the
ocean, once again (Randall 1990; Sherbondy 1992).
Symbolic of the sea, the seashell has been an essen-
tial part of rituals for centuries. Pedro Pizarro observed
that sea shells were a major ceremonial commodity
offered during rainmaking and spring-encouraging
rituals in the highlands because of the mollusks close
association with the mother water, the sea (Cobo 1990
[1653]). Spondylus, or spiny oyster shell, highly prized
for offerings, was used in jewelry by pre-Inka and Inka
ritualists. The conch shell is still used in such highland
communities as Hatun Qero, where it is blown to
begin ceremonies, as it produces a deep haunting call.
Shells provide a symbolic association with the ocean,
the mother of all qochas, and show an abiding respect
for the concept of qocha.
Wira- (vira) means fat or foam of the sea (Holguin
1952 [1608]:65, in Randall 1987), and when added to
qocha becomes Wiraqocha, or the source of all life.
Ticsi Wiraqocha (also spelled Teqsi Wiraqocha) is
then dened as the origin, or foundation. Wiraqocha
was the greatest god, the creator, a being without begin-
ning or end, who created all the other supernatural
beings, animals, men (Rowe 1946:293). Qon Teqsi
Wiraqocha traveled to Lake Titicaca from the ocean
and metaphorically gave life to the rocks, caves, and
lakes along the route. Randall (1987:70) believed this
term represented water, and that the word Wiraqocha
signified both the sun and water, but also the vital
forces that impregnated them. He further suggested
that the concept of Wiraqocha is the most difcult to
understand because the Spanish tried to make it into
an Andean equivalent of the Christian God, which was
susceptible to distortion. Wiraqocha was an invisible
god, a vital life force (1987:71). He was intimately
associated with the sea, which was also seen as the sub-
stance from which all things were created. The sea was
the source of water for all the lakes in the highlands, and
so the largest lake in the Andes, Titicaca, was revered
like the sea (Sherbondy 1992:56). The Inka origin
myth begins at Lake Titicaca, where Wiraqocha created
the sun, moon, stars, and all the ancestors. According to
popular mythology, the Inka and his sister-wife emerged
from the waters near the north end of the Island of
the Sun. This myth linked the Inkas to lakes and caves
generally and to Tiwanaku and Lake Titicaca speci-
cally, in addition to supporting stories of subterranean
water passageways.
In Ausangate the Inka myth about the recirculation
of waters and the Black Llama constellation is still told.
The Waruchiri (Huaruchiri) myth stated that the black
llama must drink the ocean waters to prevent the world
from ooding. Guaman Poma de Ayala (1980 [1615]:
254256) said that during the month of October (the
Inka month dedicated to rituals of water), the Inkas tied
black llamas in the main plaza of Hawkaypata and gave
them no drinking water in order to make them plead to
Wiraqocha for the return of the waters. These black
llamas most certainly represented the celestial Black
Llama, assuring that it would become thirsty enough
to drink the ocean waters (Randall 1990:19). In April,
when the Black Llama dark cloud constellation has its
upper culmination, it could not drink anything because
it was too high in the sky. But on October 30, just at
the beginning of the rainy season, it has its lowest cul-
mination and actually appears to be at the horizon line
(Randall 1990:19; Urton 1988). Stars setting in the west
were thought to descend into the ocean (Garcilaso de
la Vega 1961 [1609]:105). Randall believed that they
thought the waters would recycle through the Black
Llama in the sky and return back to earth in the form of
rains. Urton (1988) supported this idea that the waters
of the Vilcanota River are considered to empty into
the cosmic sea and return via the Milky Way as rains.
Bastien (1978:46,72) stated: Qoyllahuayas [inhabit-
ants of the region southeast of Lake Titicaca known as
Charasani] understand the body as a hydraulic system
circulating and distilling uids and semiuids (water, air,
blood, and food). Illness is the disruption of this cycle,
causing a gradual drying up of the body. Death is the
premature separation of the wet and dry elements.
Water symbols, in particular the qocha lake symbol,
are not merely decorative expressions according to
Quechua aesthetics. Their meaning reafrms Quechua
worldview or reality, and the interrelatedness of all aspects
of life, with water as the main focal point. In conclusion,
the sacredness of water, the concern for water storage, and
the mythological roots of this conceptual understanding
are reinforced by the persistent weaving of qocha lake
pallay in the clothing of the Quechua people in the Aus-
angate region. They reafrm their cosmology through
active use of clothing worn during rituals, including the
annual pilgrimage in the Ausangate region to El Seor
de Qoyllur Riti as they celebrate a renewal of life. The
event involves a symbolic gathering of blocks of glacial
ice, accumulated through extreme sacrice. The ice is
melted, followed by the drinking of these sacred glacial
waters. Water is celebrated in this act of renewal.
112 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Figure 7. Local dress depicting qocha designs in the CCatca market near Ausangate.
Figure 8. The ice pilgrimage and dances of Qoyllur Riti, Colqepunku, Ausangate region.
H E C K M A N 113
SHAPES OF CLOTHING AND CONTINUITY OF
EXPRESSION OF ETHNICITY
Forms of textiles and clothing shapes may divulge more
continuity through time than do designs. Clothing is
made from hand-woven cloth without cutting any of
the weaving so that no part of the labor, materials, or
artistry is wasted. The cloth is then sewn together or
separated by slits with complete nished selvages on
all four sides. It is then embellished with embroidery
or other adornment. Some clothing structures linked
to past garments have retained the same names consis-
tently since the Inka Empire for exactly the same shape
of garment. The inuence of Spanish and global cloth-
ing has changed some shapes, but others are persistently
still worn and have been for over ve hundred years.
Several Spanish chronicles documented the exact names
of Inka garments (Cobo 1990 [1653]), and again Gua-
man Poma proved to be a valuable source. Contextual
uses were noted for certain textiles such as the unkuna,
a small carrying rectangular cloth with selvages on four
sides, clearly used in the Inka period and still used today
in a ritual role for carrying the sacred coca leaf. Another
example of continuity is the Inka womans lliclla, used
open across the back with the pattern worn horizon-
tally, secured by a tupu (topo) pin in front. Contemporary
womens llicllas from Ausangateincluding Pacchanta,
Kilita, and Qero (NE) and most of the Pitumarka Val-
leyare smaller than the Inka cloth, but they are still
worn as shoulder cloth open across the back, most often
worn with the pattern displayed horizontally.
The only mens clothing items that have been
retained are the woven belt, ajota sandals, and chuspa,
the coca bag. Paradoxically the form and name of the
ajota sandals survive, but both men and women wear
ones made from recycled tire rubber today. Quechua
women have retained more textile forms from the Inka
period (for more details, see Heckman 1997, 2003).
Despite national legislation in 1780 banning the use of
Inka clothing, some remnants seem to have persisted in
the Ausangate region even though they are made today
with commercial beads and sequins.
Pacchanta and Ausangate weavers have not com-
promised their forms, the meanings of their designs,
or the aesthetics of ne weave they associate with their
ancestors. Faced with Spanish colonial, modern, and
postmodern inuences and commercial market cloth-
ing, weavers in these regions continue to value making
cloth with expressive qualities of identity. They continue
to make choices to express their ethnicity and sense of
belonging through the clothing they make and wear.
Investigations of persistent textile techniques, designs,
and forms provide challenging insights into past cos-
mology and ethnicity of ancient weavers, given that
many of the technical processes, looms, and materials
are still used today. Even when contemporary weavers
choose to incorporate synthetic yarns or materials into
their creations, the intent of message, neness of spun
yarn, and high quality of the weaving remain part of a
highly valued Andean aesthetic. Weaving continues to
be a primary expression of Andean worldview, identity,
and ethnic pride.
ACKNOWLEDGEMENTS
Portions of this chapter are excerpted from Woven Sto-
ries: Andean Textiles and Rituals (2003) by permission of
the University of New Mexico Press.
REFERENCES
Asher, M., and R. Asher
1981 Code of the quipu: A study in media, mathematics, and
culture. Ann Arbor: University of Michigan Press.
Bastien, J. W.
1978 Mountain of the condor: Metaphor and ritual in an Andean
ayllu. Prospect Heights, IL: Waveland Press, Inc.
Bawden, G.
2003 Foreword. In Woven Stories: Andean textiles and rituals
by A. Heckman, ixx. Albuquerque: University of
New Mexico Press.
Bird, J. B., and J. P. Bird
1949 Techniques. Andean Culture History, Handbook
Series 15. New York: American Museum of Natural
History.
Cereceda, V.
1986 The semiology of Andean textiles: The Talegas of
Isluga. In Anthropological history of Andean polities,
edited by J. V. Murra, N. Watchel, and J. Revel,
149173. Cambridge: Cambridge University Press.
Cobo, B.
1983 [1653] History of the Inca Empire. Translated and edited
by R. Hamilton. Austin: University of Texas Press.
1990 [1653] Inca religion and customs. Translated and edited
by R. Hamilton. Austin: University of Texas Press.
Cummings, T.
1994 Representation in the sixteenth century and the
colonial image of the Inca. In Writing without
words, alternative literacies in Mesoamerica and the
Andes, edited by E. H. Boone and W. J. Mignolo,
188219. Durham, NC: Duke University Press.
114 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Donnan, C.
1978 Moche art of Peru: Pre-Columbian symbolic communi-
cations. Los Angeles: Museum of Cultural History,
University of California.
Flores-Ochoa, J.
1968 Pastoralists of the Andes: The alpaca herders of Paratia.
Philadelphia: Institute for the Study of Human
Issues.
Garcilaso de la Vega
1961 [1609] Royal commentaries of the Incas. Orion Press,
Inc.
Gow, D. D.
1976 The gods and social change in the high Andes.
Ph.D. dissertation, Department of Anthropology,
University of Wisconsin.
Guaman Poma de Ayala, F.
1980 [1615] Primer nueva coronica y buen gobierno. Mexico:
Siglo Vientiuno.
Heckman, A. M.
1997 Quechua cultural continuity and change as reected
by the traditional textiles and clothing of Ausan-
gate, Peru. Ph.D. dissertation, Department of Latin
American Studies, University of New Mexico.
2003 Woven stories: Andean textiles and rituals. Albuquer-
que: University of New Mexico Press.
Lumbreras, L. G.
1974 The peoples and cultures of ancient Peru. Translated
by B. J. Meggers. Washington, D.C.: Smithsonian
Institution Press.
Moseley, M. E.
1992 The Incas and their ancestors. London: Thames and
Hudson.
Murra, J. V.
1962 Cloth and its functions in the Inca state. American
Anthropologist 64 (4): 710727.
1980 Incas. Public Broadcasting Associates, PBS and the
Pacic Arts. Los Angeles: Nesmith Enterprises. 58
minutes. Videocassette.
Ortiz, A.
1994 The dynamics of Pueblo cultural survival. In North
American Indian anthropology, edited by R. J. Demal-
lie and A. Ortiz, 296306. Norman: University of
Oklahoma Press.
Paul, A.
1990 Paracas ritual attire: Symbols of authority in ancient
Peru. Norman: University of Oklahoma Press.
Randall, R.
1982 Qoyllur Riti, an Inca esta of the Pleiades: Reections
on time and space in the Andean world. Boletn del
Instituto Frances de Estudios Andinos 11. Lima.
1987 Del tiempo y del rio. Boletn de Lima 54 (November):
6995.
1990 The myth story of Kuri Qoyllur. Journal of Latin
American Lore 16 (1): 345.
Rowe, A. P.
1977 Warp-patterned weaves of the Andes. Washington,
D.C.: The Textile Museum.
Rowe, J. H.
1946 Inca culture at the time of the Spanish con-
quest. In Handbook of South American Indians, 2:
183330. Washington, D.C.: Bureau of American
Ethnology.
1979 Standardization in Inca tapestry tunics. In A. P.
Rowe, E. P. Benson, and A.-C. Schaffer, eds. The
Junios B. Bird Pre-Columbian Textile Conference,
May 1920, 1973, 239264, Washington D.C.
Sallnow, M. J.
1987 Pilgrims of the Andes. Washington, D.C.: Smithson-
ian Institution Press.
Sawyer, A. R.
1963 Tiahuanaco tapestry design. The Museum of Primitive
Art Studies 3. Washington, D.C.: The Museum of
Primitive Art.
Sherbondy, J. E.
1992 Water ideology in Inca ethnogenesis. In Andean
cosmologies through time: Persistence and emergence,
edited by R. V. H. Dover, K. E. Seibold, and J. H.
McDowell, 166201. Bloomington: University of
Indiana Press.
Stone-Miller, R.
1992 To weave for the sun: Ancient Andean textiles. London:
Thames and Hudson.
Urton, G.
1988 At the crossroads of the earth and the sky: An Andean
cosmology. Austin: University of Texas Press.
Weiner, A. B., and J. Schneider
1991 Cloth and human experience. Washington, D.C:
Smithsonian Institution Press.
115
T
here are multiple ways one might examine clothing
styles as a reection of possible ethnic identity in
Andean contexts. Within contemporary communi-
ties, the work of Meisch (1986, 1996) in Tarabuco
and Saraguro, Zorn (1995) in Sacaca and Taquile,
Medlin (1996) in Calcha, Seibold (1992, 1995) in
Choquecancha, and others is valuable in discuss-
ing the ways that clothing style reects indigenous
identity, albeit, in constant change. Anyone who has
lived for some time in the highlands of Bolivia or Peru
becomes accustomed to the immediate identication
of community-specic garments, and the variations
as well. Tarabucano men and women are clearly iden-
tied from the region near Sucre, Bolivia, through
their red-and-yellow-striped hand-woven mantles,
ponchos, and skirts (gure 1). They wear an abbrevi-
ated costume for work and create elaborate additions
for festivals, all within a specic ethnic style recogniz-
ing identity in connected, but distinct layers. Outsiders
understand the basic costume as a garment style while
insiders understand deep meaning expressed in cloth-
ing. With persistence and skill, anthropologists are
able to mine this deeper meaning. Consider Seibolds
(1995:324325) observation of blue stripes in a group
of newly woven womens textiles. She had lived and
worked among the people of Choquecancha in the
highlands of Peru for a decade and knew that, worn
by women of child-bearing age, these normally green
stripes represented the fertile earth. In this commu-
nity, girls before the age of menstruation wear only
the brown of the unplowed and unproductive while
married women weave bright stripes into their textiles,
especially brilliant, verdant, green. Seibold had never
seen blue stripes woven and worn by women of mar-
riageable age and she didnt receive any convincing
answers after much questioning. Finally a few young
informants admitted that women now wearing intra-
uterine devices had changed their costume to reect
their new status: women are wearing their fertility,
or lack thereof, like badges on their sleeves (Seibold
1995:325). Could textile patterning have provided such
intimate information in archaeological contexts?
CHAP T E R 8
North Coast Style after Moche:
CL OT HI NG AND I DE NT I T Y
AT E L BRUJ O, CHI CAMA
VAL L E Y, P E RU
d
bp n bp n bp n bp n
168 22 0.06 175 1 0.00 189 12 0.03 204 49 0.14 171 26 0.08
172 16 0.05 179 88 0.26 193 169 0.49 206 29 0.08 175 45 0.13
176 149 0.43 183 10 0.03 197 144 0.42 208 16 0.05 179 74 0.22
180 113 0.33 187 57 0.17 205 19 0.06 210 206 0.60 181 5 0.02
184 44 0.13 191 61 0.18 212 23 0.07 183 157 0.46
195 38 0.11 214 1 0.00 185 13 0.04
199 87 0.25 216 22 0.06 187 11 0.03
203 2 0.01 191 1 0.00
195 8 0.02
D17S933 D18S535 D19S253 D21S1435 D22S684
H=0.67 H=0.77 H=0.73 H=77 H=0.72
bp n bp n bp n bp n bp n
188 4 0.01 144 78 0.22 212 11 0.03 170 84 0.25 231 101 0.29
190 20 0.06 148 55 0.16 224 69 0.2 174 84 0.25 235 113 0.33
192 76 0.22 152 129 0.37 228 94 0.28 178 100 0.30 239 94 0.27
196 175 0.51 156 84 0.24 232 126 0.37 182 14 0.04 243 37 0.11
198 26 0.08 236 34 0.1 186 38 0.11 255 1 0.00
202 18 0.05 240 6 0.02 190 18 0.05
204 26 0.08
206 1 0.00
a
heterozygosity value, the percentage of individuals likely to be heterozygous at this locus
b
length of alleles in base pairs (bp)
c
number of alleles observed
d
frequency of occurrences of allele
Table 2. Mitochondrial DNA haplotypes.
1
6
1
0
4
1
6
1
7
9
1
6
1
8
3
1
6
1
8
9
1
6
2
1
5
1
6
2
1
7
1
6
2
2
3
1
6
2
8
8
1
6
2
9
4
1
6
2
9
8
1
6
3
0
2
1
6
3
1
1
1
6
3
2
5
1
6
3
2
7
1
6
3
5
3
1
6
3
6
2
reference
1
C C A T A T C T C T A T T C C T n frequency
B
2
1 . . C C . C . . . . . . . . . . 75 0.48
2 . . C C . C . C . . . C . . . . 11 0.07
3 . . C C G C . . . . . . . . . . 1 0.006
C 1 . . . . . . T . . C . . C T . . 49 0.32
D 1 G . . . . . . . T . G . C . . C 13 0.08
2 . A . . . . T . . . . . C . T C 6 0.04
1
Cambridge reference sequence (Anderson et al. 1981). The numbers above it refer to nucleotide positions.
2
Haplogroup to which the sequence belongs (A, B, C, or D)
144 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Table 3. Nuclear STRP and mtDNA F-statistics analog results for males and females from four villages.
Among villages Among sexes within villages Within sexes
STRPs Variance % Total
a
ct
b
p
c
Variance % Total
sc
p Variance % Total
st
p
combined 0.231 3.90 0.039 * -0.035 -0.59 0.000 * 5.718 96.70 0.033 *
Among villages Among sexes within villages Within sexes
Variance % Total
ct
p Variance % Total
sc
p Variance % Total
st
p
mtDNA 0.370 15.57 0.156 * 0.136 5.72 0.068 ~ 1.872 78.70 0.213 *
a
percent of total variance found in this variance component. Both variance components and percents can be negative (see text).
b
F-statistic analogue
c
probability that a more extreme value could be generated by chance. * indicates that p<0.05. ~ indicates that p<0.06
Table 4. Genetic distances between males and
females in each village.
STRPs mtDNA
village 1 -0.01 -0.03
village 2 -0.01 0.15*
village 4 0.01 -0.08
village 5 0.00 0.09*
*signicant at p<0.05 level
Table 5. Lane and Sublett method for assessing postmarital residence change.
Marker female F
st
male F
st
female F
st/
male F
st
p
1
STRP 0.036 0.029 0.036 = 1.24 0.71 not signicant
0.029
mtDNA 0.225 0.181 0.225 = 1.24 0.42 not signicant
0.181
1
probability that a more extreme value could occur by chance.
Table 6. Spence method of assessing postmarital residence change.
STRP
Village n
1
(males) n (females) V
m
/V
f
2
1 24 15 1.005
2 25 28 0.985
4 10 12 0.977
5 16 15 1.069
mtDNA
Village n (males) n (females) V
m
/V
f
1 24 15 0.994
2 25 28 1.043
4 10 12 1.302
5 16 15 3.292*
1
n is the sample number
2
V
m
is the male variance, V
f
is the female variance
*greater than F
.05
value
Table 7. Methods of Lane and Sublett (1972) and Spence (1974) predictions for intergender
differences in nuclear and mitochondrial genetic variance.
Patrilocal Residence Matrilocal Residence Endogamy
NUCLEAR DNA
Lane and Sublett Method V
1
female
/V
male
< 1.0 V
female
/V
male
> 1.0 V
female
/V
male
= 1.0
Spence Method V
female
/V
male
> 1.0 V
female
/V
male
< 1.0 V
female
/V
male
= 1.0
MITOCHONDRIAL DNA
Lane and Sublett Method V
female
/V
male
= 1.0 V
female
/V
male
> 1.0 V
female
/V
male
= 1.0
Spence Method V
female
/V
male
= 1.0 V
female
/V
male
< 1.0 V
female
/V
male
= 1.0
1
variance
W I L L I A M S 145
Furthermore, only part of the mitochondrial genome
was sequenced, so sequencing a different section might
have produced a slightly different result.
The Lane & Sublett (table 5) and the Spence (table 6)
methods were used to determine whether the observed
effects could be attributed to the greater mobility of
women among Yanomam villages. Neither method
showed any real signicant differences
7
in mobility pat-
terns, implying that the general pattern of endogamy
observed in Yanomam villages is not significantly
inuenced by the small amount of gene ow known to
occur in this group. MtDNA analyses, unlike nuclear
DNA analyses, however, do not discriminate between
situations where women move (patrilocal residence)
and situations where no one moves (endogamy) (table
7). Consequently, the results of the mtDNA residence
pattern analyses are consistent with either endogamy
or patrilocal residence. Therefore, the observed differ-
ences between nuclear and mtDNA could be the result
of gene ow having greater impact on mtDNA patterns,
but currently available analytical methods could not dis-
tinguish between the two possibilities.
DISCUSSION
The settlement history of the Andes is relatively short,
at most 15,00020,000 years long (Dillehay 1997; Melt-
zer et al. 1997), so members of ethnic groups in this area
are likely to have diverged relatively recently from a
common ancestral population. The amount of diversity
these original settlers brought with them is still a matter
of debate (Merriwether, Rothhammer, and Ferrell 1995;
Schurr, Ballinger, and Gan 1990; Torroni et al. 1993;
Wallace and Torroni 1992) and ultimately will be best
addressed through ancient DNA analyses. This recent
divergence date means that Andean groups, even those
who belong to different ethnic groups, are likely to be
quite closely related to each other, so the genetic pat-
terns observed among the Yanomam are a reasonable
approximation of genetic patterns likely to be observed
among the more closely related groups.
The basic question addressed here was whether
genetic studies in general, and mtDNA studies in par-
ticular, could be successfully used to identify genetic
relationships among groups of relatively closely related
individuals such as would likely be found in nearby
archaeological sites. I used the Yanomam as a test
case because good genealogical information was avail-
able to minimize the potentially confounding affects of
performing genetic analyses on individuals of unknown
biological relationship to each other. The Yanomam
data was used to address some fundamental questions
that included whether any genetic differences could
be identied among groups as closely related to each
other as the individuals who lived in these villages, and
especially, whether genetic variation in mtDNA, which
has some unique properties, could identify differences
in these groups.
The nuclear microsatellite analyses were an excellent
indicator of endogamy in closely related groups like the
Yanomam Shamatari villages described here. Whether
they are the best means of identifying endogamy in the
prehistoric past is less clear because of the limitations
of using a damaged template like ancient DNA. The
presence of only a single copy of nuclear DNA per cell
means that even when amplication (or copy making) is
successful, often only one of the two alleles present in
a persons nuclear genome will amplify. Zierdt and col-
leagues (1996) illustrated the potential problems involved
in the use of nDNA markers in ancient DNA studies.
Twice as many individuals in their study were typed
as homozygous at the VWA31/A microsatellite loci as
would be expected under Hardy Weinberg Equilibrium
conditions. While similarly high levels of homozygosity
might be observed as a consequence of inbreeding, they
concluded that the observed values were more likely
caused by technical problems resulting from the failure
to amplify both alleles present in some samples. Inated
estimates of homozygosity resulting from unequal allele
amplication will result in incorrect allele frequencies,
making nuclear DNA microsatellites a poor choice
for population genetic studies of prehistoric groups
until techniques have been developed to ensure greater
genotype accuracy. For now, nuclear DNA microsatellite
studies based on ancient DNA are better used in studies
requiring only a few individuals, like parent-offspring
determinations, where the number of typings per locus
is small, so the number of loci can be increased easily.
Microsatellite markers that give homozygous typings in
any of the study individuals can then be discarded and
replaced by another marker, a strategy highly impractical
to implement in larger studies.
The research described here shows that signicant
mtDNA haplotypes can be present even in closely
related groups. The study provides an estimate of the
kind of variation that might be expected to be observed
in individuals buried at a single site or in a group of
closely related sites, which makes artificially high
mtDNA haplotype frequencies in aDNA studies caused
by template degradation more likely to be detected.
146 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Although mtDNA will remain the preferred genome
for study in ancient DNA research, it should be noted
that the Yanomam mtDNA analyses failed to fulll
several critical expectations for the identication of
the relationships among endogamous groups with a
recent common ancestor. The mtDNA genetic dis-
tances did not match the ssion history of the villages
or the distances calculated from nuclear DNA, possibly
because of undetected gene ow or sampling problems.
Although the Lane & Sublett and Spence models per-
formed reasonably well, both matching the expectations
for genetic relationships among largely endogamous
groups, they did not provide information that could be
used to choose between gene ow and sampling as pos-
sible explanations for the discrepancies in the mtDNA
patterns. Neither model could detect whether low lev-
els of sex-mediated gene ow inuenced the mtDNA
patterns. Their value to archaeological research is
even more problematic because the models require
assumptions that are difcult to meet in archaeologi-
cal studies (Konigsberg 1987, 1988; Souza Aguiar and
Neves 1991; Williams-Blangero and Blangero 1990).
Konigsberg (1987, 1988) noted that sites must overlap
in time to belong to the same mating network, and
that only groups who participated in the same mat-
ing network can be included when using the model,
which requires exacting dating techniques and access to
information not commonly available to archaeologists.
Williams-Blangero and Blangero (1990) speculated that
the disappointing results they obtained when using the
Spence model to study a modern group was caused
by unequal genetic variance among the villages they
included in their work. The Lane & Sublett model can
work well, but requires assumptions that are difcult to
meet when using archaeological samples. The Spence
model is such a weak statistical test that it may not work
under the best of conditions in living populations.
If the mtDNA patterns observed in the Yanomam
accurately reect a degree of gene ow undetected by
nuclear DNA, Y haplotype analyses may provide an
important additional source of information. Bamshad
and colleagues (1998) study of gene ow among Indian
castes effectively demonstrated the advantages of com-
bining mtDNA and Y haplotype frequency analyses.
Y haplotyping provides an additional locus and adds a
measure of male relatedness to complement the measure
of female relationships provided by mtDNA. Y chro-
mosome studies are an appropriate choice for ancient
DNA studies because, although the DNA will be pres-
ent in smaller amounts than mtDNA, uneven allele
amplication will not affect Y chromosome analyses.
Each male has only one Y chromosome, so no product
forms if the allele fails to amplify. The introduction of
error into subsequent analyses can be avoided because
missing or null alleles are immediately apparent.
Finally, combining aDNA analyses with other
techniques that measure similar, but slightly different
biological processes, such as traditional biodistance and
stable isotope studies, will strengthen the conclusions
that can be drawn from each. Traditional biodistance
studies measure metric and non-metric trait variation.
These phenotypic or epigenetic traits are physical
expressions of combinations of alleles at multiple
genetic loci, but their expression is also affected by
developmental and environmental processes. Stable iso-
tope analyses measure the ratios of naturally occurring,
nonradioactive isotopes of elements such as strontium
(Sr), oxygen (O), and lead (Pb) in human bone or
tooth samples (Katzenberg 2000; Price, Burton, and
Bentley 2002). The concentrations of each elements
stable isotopes, and thus their ratios, vary throughout
the physical environment and are incorporated into
the tissues of the people who live in the area through
their diet. Stable isotopes yield information about the
physical movements of individuals during their lives by
comparing the ratios in teeth that result from child-
hood diet with the ratios in bone that reect adult diet
(Bentley, Chikhi, and Price 2003; Bentley, Price, and
Lning 2002; Bentley et. al 2002; Katzenberg 2000;
Price, Grupe, and Schrter 1998). Both of these tech-
niques provide important information either about the
biological relationships among individuals or of popula-
tion movements across the landscape that can reinforce
data from aDNA studies.
CONCLUSION
In general, the longer the separation time between
groups and the more severe the reduction in gene ow,
the greater the genetic divergence will be and the easier
it will be for current methods to discriminate between
groups. The following example serves as a reminder
that ethnic groups are not necessarily biologically
identiable entities and that shared culture is often
the dening characteristic of ethnicity. In a previous
study of blood proteins (Chagnon et al. 1970), geneti-
cists were surprised to nd a Yanomam village that
more closely resembled the Makiritare, a neighbor-
ing tribe whose nearest village was some 200 miles to
the north of this particular Yanomam village. The
W I L L I A M S 147
Makiritare and Yanomam speak different languages,
and their cultural practices differ in many ways (Cha-
gnon et al. 1970). Ethnographers eventually discovered
that the ancestors of these villagers had lived on the
Makiritare-Yanomam frontier around 1900, and the
earlier village had included a half-Makiritare headman
and several Makiritare women who had been abducted
by men from this village. Within three generations,
some blood protein alleles previously found only in
the Makiritare had reached higher frequencies in this
village than in any of the Makiritare villages surveyed
(Chagnon et al. 1970:347). The villagers considered
themselves Yanomam, not a village of mixed Yano-
mam and Makiritare ancestry. This case highlights the
rapidity with which signicant gene ow can ameliorate
genetic differences between two groups who perceive
each other as different ethnic groups.
The amount and pattern of gene ow among neigh-
boring ethnic groups will dramatically affect the extent
to which ethnic groups can be distinguished by dif-
ferences in their genetic patterns. Because ethnicity
is self-dened, individuals may consider themselves
members of a separate ethnic group, even when exten-
sive intergroup gene ow, whether sanctioned or not,
occurs with their neighbors. Therefore, ethnicity will
have variable effects on allele frequencies, making their
use as ethnic identiers similarly complicated. While
some ethnic groups may be completely closed (endoga-
mous groups), other ethnic groups may have few rules
concerning marriage to individuals outside the group.
In these latter cases, there may be no genetic correlate
of shared ethnicity. In reality, these two extremes of
behavior likely will be rare. More likely, patterned
gene ow will result in differences in allele frequencies
among ethnic groups, but the effects of these patterns
may be difcult to interpret.
This variability means that ethnic groups can never
be identied solely from biological attributes. Often,
putative prehistoric ethnic groups can be dened by
shared cultural traits, such as textile production or
mortuary custom, with genetic evidence used to test
whether the individuals that share these characteristics
share a close biological relationship as well. Sometimes,
the cultural and biological patterns will correspond,
as Blom and colleagues (1998) demonstrated when
they found signicant biological differences between
individuals buried at Chiribaya and Tiwanaku sites,
two neighboring ethnic groups who once lived in the
Osmore Drainage. Elsewhere, groups may share cul-
tural characteristics, but not biology. For example, after
Inka conquest, Wanka elite houses began to resemble
Inka houses (Costin and Earle 1989). Ethnohistoric
documents showed that these homes continued to
belong to Wanka elite, so the researchers could focus on
explaining the reasons for this change. Costin and Earle
reasoned that the Wanka elite adopted Inka-style archi-
tecture to emphasize their association with Inka rulers
and emphasize their own power within the community.
Had no ethnohistoric documents existed, researchers
might have incorrectly concluded that this change in
architectural style signaled the arrival of an Inka elite as
a means of direct rule in this area. Sutters (2000) dental
trait analysis provides an example of the opposite case,
where biology is shared, but cultural characteristics differ.
His study indicates that, although cultural traits change
through time in the Azapa Valley, biological continuity
is present. Clearly, the combined use of culture traits and
biological markers can be a powerful means to identify
and study prehistoric ethnic groups.
MtDNA has already proven useful in the study of
continental levels of variation and should help clarify
regional and local population histories as well. The fact
that mtDNA diversity exists in small, closely related
populations indicates that prehistoric mtDNA popula-
tion studies are feasible. The combination of mtDNA
and Y chromosome analyses should provide a powerful
means with which to identify gene ow in prehistory.
The biological and social nature of the relation-
ships that existed among Andean peoples are of great
interest to archaeologists, but are extremely difcult to
discern in the archaeological record. No one method
is destined to provide the complete answer. Combining
the information gained from biological and chemical
techniques with careful studies of material culture will
be required to unravel the intricate interactions among
prehistoric peoples and enable us to better understand
the importance of ethnicity in regional culture history
and evolution.
NOTES
1. Base pairs are complementary pairs of nucleic acids whose
bonds with each other form the rungs of the DNA helix.
2. Some variation may be present in the mitochondrial
genomes of individuals, but will have relatively little impact
on studies of this kind.
3. Restriction enzymes recognize and cut unique four-to-
six-bases-long DNA sequences. Not all individuals have the
same sequence in these regions, so some individuals will have
the restriction site, while others will not.
148 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
4. There are two hypervariable sections within the d-loop or
control region of the mitochondrial genome. The d-loop does
not code for any genes, it is where transcription (genome duplica-
tion) initiates, so the region is more variable among individuals.
The hypervariable regions are areas within the d-loop where the
largest amount of nucleotide variation is concentrated.
5. Allele frequencies, also referred to as gene frequencies,
usually refer to frequencies of various forms of a gene or small
DNA segment. Haplotype frequencies are used when refer-
ring to alternative mtDNA sequences because the genome is
inherited as a single unit composed of multiple genes.
6. A heterozygosity value is the percentage of individuals
likely to have two different alleles, or genetic sequences at a
particular locus.
7. The mitochondrial DNA intersex comparison for village
5 indicated that male variance was signicantly larger than
female variance.
REFERENCES
Aldenderfer, M. S., ed.
1993 Domestic architecture, ethnicity, and complementarity in
the south-central Andes. Iowa City: University of Iowa
Press.
Anderson, A., A. T. Bankier, B. G. Barrell, M. H. L. de
Bruijn, A. R. Coulson, J. Drouin, I. C. Eperon, D.
P. Nierlich, B. A. Roe, F. Sanger, P. H. Schreier, A.
J. H. Smith, R. Staden, and I. G. Young
1981 Sequence and organization of the human mitochon-
drial genome. Nature 290:457470.
Bamshad, M. J., W. S. Watkins, M. E. Dixon, L. B. Jorde, B.
B. Rao, J. M. Naidu, B. V. R. Prasad, A. Rasanaya-
gam, and M. F. Hammer
1998 Female gene ow straties Hindu castes. Nature
395:651652.
Bawden, G.
1993 An archaeological study of social structure and eth-
nic replacement in residential architecture of the
Tumilaca Valley. In Domestic architecture, ethnicity,
and complementarity in the south-central Andes, edited
by M. S. Aldenderfer, 4254. Iowa City: University
of Iowa Press.
Bentley, R. A., L. Chikhi, and T. D. Price
2003 The Neolithic transition in Europe: Comparing
broad scale genetic and local scale isotopic evidence.
Antiquity 77 (295): 6366.
Bentley, R. A., T. D. Price, and J. Lning
2002 Prehistoric migration in Europe: Strontium iso-
tope analysis of early Neolithic skeletons. Current
Anthropology 43 (5): 799804.
Bentley, R. A., T. D. Price, J. Lning, D. Gronenborn, J.
Wahl, and P. D. Fullagar
2002 Human migration in early Neolithic Europe. Cur-
rent Anthropology 43:799804.
Braud-Colomb, E., R. Roubin, J. Martin, N. Maroc, A.
Gardeisen, G. Trabuchet, and M. Goossens
1997 Reply to Cooper. American Journal of Human Genet-
ics 60:10021003.
Blom, D. E.
1999 Tiwanaku regional interaction and social identity:
A bioarchaeological approach. Ph.D. disserta-
tion, Department of Anthropology, University of
Chicago.
Blom, D. E., B. Hallgrimsson, L. Keng, M. C. Lozada-Cerna,
and J. E. Buikstra
1998 Tiwanaku colonization: Bioarchaeological impli-
cations for migration in the Moquegua Valley, Peru.
World Archaeology 30 (2): 238261.
Cano, R. J., H. N. Poinar, N. J. Peiniazek, A. Acra, and G.
O. Poinar, Jr.
1993 Amplication and sequencing of DNA from a 120
135-million-year-old weevil. Nature 363:536538.
Carlyle, S. W., R. L. Parr, M. G. Hayes, and D. H.
ORourke
2000 Context of maternal lineages in the greater South-
west. American Journal of Physical Anthropology 113
(1): 85101.
Chagnon, N. A.
1966 Yanomam warfare, social organization, and mar-
riage alliances. Ph.D. dissertation, Department of
Anthropology, University of Michigan.
1968a Yanomam: The erce people. New York: Holt, Rine-
hart and Winston.
1968b The culture-ecology of shifting (pioneering) culti-
vation among the Yanomam Indians. Proceedings
of the VIII international congress of anthropological
and ethnological sciences 3:249255. Tokyo: Science
Council of Japan.
1974 Studying the Yanomam. New York: Holt, Rinehart
and Winston.
1975 Genealogy, solidarity, and relatedness: Limits to
local group size and patterns of ssioning in an
expanding population. Yearbook of Physical Anthro-
pology 19:95110.
1979 Mate competition, favoring close kin, and vil-
lage ssioning among the Yanomam Indians. In
Evolutionary biology and human social behavior: An
anthropological perspective, edited by N. A. Chagnon
and W. Irons, 86132. North Scituate, MA: Dux-
bury Press.
W I L L I A M S 149
1997 Yanomam. The erce people. 5th ed. NewYork: Holt,
Rinehart and Winston, Inc.
Chagnon, N. A., and W. Irons, eds.
1979 Evolutionary biology and human social behavior: An
anthropological perspective. North Scituate, MA:
Duxbury Press.
Chagnon, N. A., J. V. Neel, L. Weitkamp, H. Gershowitz,
and M. Ayres
1970 The inuence of cultural factors on the demography
and pattern of gene ow from the Makiritare to the
Yanomama Indians. American Journal of Physical
Anthropology 32:339350.
Cooper, A.
1997 Reply to Stoneking: Ancient DNAHow do you
really know when you have it? American Journal of
Human Genetics 60:10011002.
Costin, C. L., and T. Earle
1989 Status distinction and legitimization of power as
reected in changing patterns of consumption in
late prehispanic Peru. American Antiquity 54 (4):
691714.
Dillehay, T. D., ed.
1997 Monte Verde, a late Pleistocene settlement in Chile. Vol-
ume 2: The archaeological context: An interpretation.
Washington, D.C.: Smithsonian Institution Press.
Doran, G. H., D. N. Dickel, W. E. Ballinger Jr., O. F. Agee,
P. J. Laipsis, and W. W. Hauswirth
1986 Anatomical, cellular, and molecular analysis of
8,000-yr-old human brain tissue from the Windover
archaeological site. Nature 323:803806.
Droessler, J.
1981 Craniometry and biological distance: Biocultural continuity
and change at the late Woodland Interface. Research Series
1. Evanston, IL: Center for American Archaeology.
Evershed, R. P., H. A. Bland, P. F. van Bergen, J. F. Carter, M.
C. Horton, and P. A. Rowley-Conway
1997 Volatile compounds in archaeological plant remains
and the Maillard reaction during decay of organic
matter. Science 278:432433.
Ewing, B., and P. Green
2000 Analysis of expressed sequence tags indicates 35,000
human genes. Human Genetics 25:232234.
Excofer, L., P. E. Smouse, and J. M. Quattro
1992 Analysis of molecular variance inferred from metric
distances among DNA haplotypes: Application to
human mitochondrial DNA restriction data. Genetics
131:479491.
Ferguson, R. B.
1995 Yanomami warfare: A political history. Santa Fe, NM:
School of American Research Press.
Hnni, C., T. Brousseau, V. Laudet, and D. Stehelin
1995 Isopropanol precipitation removes PCR inhibitors
from ancient bone extracts. Nucleic Acids Research 23
(5): 881882.
Hauswirth, W. W.
1994 Ancient DNA: An introduction. Experientia 50:
521523.
Hauswirth, W. W., C. D. Dickel, D. J. Rowold, and M. A.
Hauswirth
1994 Inter- and intrapopulation studies of ancient
humans. Experientia 50:585591.
Henley, P.
1995 Yanomami: Masters of the spirit world. San Francisco:
Chronicle Books.
Hss, M., P. Jaruga, T. H. Zastawny, M. Dizdaroglu, and S.
Pbo
1996 DNA damage and DNA sequence retrieval from
ancient tissues. Nucleic Acids Research 24 (7): 1304
1307.
Kaestle, F.
1997 Molecular analysis of ancient Native American
DNA from western Nevada. Nevada Historical Soci-
ety Quarterly 40 (1): 8596.
Kaestle, F. A.
1995 Evidence for the identity of the descendants of the pre-
historic Stillwater Marsh population. In Bioarchaeology
of the Stillwater Marsh: Prehistoric human adaptation in
the western Great Basin, edited by C. S. Larsen and R.
L. Kelly, 7380. Anthrological Papers of the American
Museum of Natural History 77. New York.
Kaestle, F. A., and K. A. Horsburgh
2002 Ancient DNA in anthropology: Methods, applica-
tions, and ethics. Yearbook of Physical Anthropology 45:
92130.
Kaestle, F. A., J. G. Lorenz, and D. G. Smith
1999 Molecular genetics and the Numic expansion. In
Prehistoric lifeways in the Great Basin wetlands: Bio-
archaeological reconstruction and interpretation, edited
by B. E. Hemphill and C. S. Larsen, 167183. Salt
Lake City: University of Utah Press.
Kaestle, F. A., and D. G. Smith
2001 Ancient mitochondrial DNA evidence for prehis-
toric population movement: The Numic expansion.
American Journal of Physical Anthropology 115 (1):
112.
Katzenberg, M. A.
2000 Stable isotope analysis: A tool for studying past diet,
demography, and life history. In Biological anthropol-
ogy of the human skeleton, edited by M. A. Katzenberg
and S. Saunders, 305327. New York: Wiley-Liss.
150 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Kolman, C. J., and N. Tuross
2000 Ancient DNA analysis and human populations.
American Journal of Physical Anthropology 11 (1):
523.
Konigsberg, L. W.
1987 Population genetic models for interpreting pre-
historic intra-cemetery biological variation. Ph.D.
dissertation, Department of Anthropology, North-
western University.
1988 Migration models of prehistoric postmarital resi-
dence. American Journal of Physical Anthropology 77:
471482.
Lawlor, D. A., C. D. Dickel, W. W. Hauswirth, and P. Par-
ham
1991 Ancient HLA genes from 7,500-year-old archaeo-
logical remains. Nature 349:785788.
Lewin, B.
1994 Genes V. Oxford: Oxford University Press.
Lindahl, T.
1993 Instability and decay of the primary structure of
DNA. Nature 362:709715.
Lizot, J.
1985 Tales of the Yanomami: Daily life in the Venezuelan for-
est. Cambridge: Cambridge University Press.
Lozada-Cerna, M. C.
1998 The seoro of Chiribaya: A bio-archaeological
study in the Osmore Drainage of southern Peru.
Ph.D. dissertation, Department of Anthropology,
University of Chicago.
MacCluer, J. W., J. V. Neel, and N. A. Chagnon
1971 Demographic structure of a primitive population: A
simulation. American Journal of Physical Anthropology
35:193208.
Malhi, R. S., and D. G. Smith
2002 Brief communication: Haplogroup X conrmed
in prehistoric North America. American Journal of
Physical Anthropology 119 (1): 8486.
Meltzer, D. J., D. K. Grayson, G. Ardila, A. W. Barker, D.
F. Dincause, C. V. Haynes, F. Mena, L. Nunez, and
D. Stanford
1997 On the Pleistocene antiquity of Monte Verde, south-
ern Chile. American Antiquity 62:659663.
Merriwether, D. A., F. Rothhammer, and R. E. Ferrell
1994 Genetic variation in the New World: Ancient teeth,
bone, and tissue as sources of DNA. Experientia 50:
592601.
1995 Distribution of the four founding lineage haplo-
types in Native Americans suggests a single wave of
migration for the New World. American Journal of
Physical Anthropology 98:411430.
Neel, J. V.
1967 The genetic structure of primitive human popula-
tions. Japanese Journal of Human Genetics 12:116.
1970 Lessons from a primitive people. Science 170:
815822.
Neel, J. V., T. Arends, C. Brewer, N. A. Chagnon, H. Ger-
showitz, M. Layrisse, Z. Layrisse, J. MacCluer, E.
Migliazza, W. Oliver, F. M. Salzano, R. Spielman,
R. H. Ward, and L. Weitkamp
1972 Studies on the Yanomama Indians. In Human
genetics: Proceedings of the 4th international congress
of human genetics, Paris, September 1971, 96111.
Amsterdam: Excerpta Medica.
Neel, J. V., F. Rothhammer, and J. C. Lingoes
1974 The genetic structure of a tribal population, the
Yanomama Indians. X. Agreement between rep-
resentations of village distances based on different
sets of characteristics. American Journal of Human
Genetics 26 (3): 281303.
Neel, J. V., and R. H. Ward
1972 The genetic structure of a tribal population, the
Yanomama Indians VI. Analysis by F-statistics
(including a comparison with the Makiritare and
Xavante). Genetics 73:639666.
Neel, J. V., and K. M. Weiss
1975 The genetic structure of a tribal population, the
Yanomama Indians. XII. Biodemographic studies.
American Journal of Physical Anthropology 42:2552.
Nielsen, H., J. Endberg, and I. Thuesen
1994 DNA from Arctic human burials. In Ancient DNA:
Recovery and analysis of genetic material from paleon-
tological, archaeological, museum, medical, and forensic
specimens, edited by B. Herrmann and S. Hummel,
122140. New York: Springer-Verlag.
ORourke, D. H., S. W. Carlyle, and R. L. Parr
1996 Ancient DNA: Methods, progress, and perspectives.
American Journal of Human Biology 8:557571.
ORourke, D. H., R. L. Parr, G. Hayes, and S. W. Carlyle
1999 Molecular genetic variation in prehistoric inhabit-
ants of the eastern Great Basin. In Prehistoric lifeways
in the Great Basin wetlands: Bioarchaeological recon-
struction and interpretation, edited by B. E. Hemphill
and C. S. Larsen, 84102. Salt Lake City: University
of Utah Press.
Pbo, S.
1985a Molecular cloning of ancient Egyptian mummy
DNA. Nature 314:644645.
1985b Preservation of DNA in ancient Egyptian mummies.
Journal of Archaeological Science 12:411417.
1989 Ancient DNA: Extraction, characterization,
W I L L I A M S 151
molecular cloning, and enzymatic amplication.
Proceedings of the National Academy of Sciences USA
86:19391943.
Parr, R. L., S. W. Carlyle, and D. H. ORourke
1996 Ancient DNA analysis of Fremont Amerindians of
the Great Salt Lake wetlands. American Journal of
Physical Anthropology 99:507518.
Price, T. D., R. A. Bentley, D. Gronenborn, J. Lning, and
J. Wahl
2001 Human migration in the Linearbandkeramik of
central Europe. Antiquity 75:593603.
Price, T. D., J. H. Burton, and R. A. Bentley
2002 The characterization of biologically available
strontium isotope ratios for the study of prehistoric
migration. Archaeometry 44 (1): 117135.
Price, T. D., G. Grupe, and P. Schrter
1998 Migration and mobility in the Bell Beaker period in
central Europe. Antiquity 72:405411.
Ramos, A. R.
1995 Sanum Memories: Yanomami ethnography in times of
crisis. Madison: University of Wisconsin Press.
Ribeiro-Dos Santos, A. K. C., S. E. B. Santos, A. L. Machado,
V. Guapindaia, and M. A. Zago
1996 Heterogeneity of mitochondrial DNA haplotypes
in pre-Columbian Natives of the Amazon region.
American Journal of Physical Anthropology 101:
2937.
Richards, M. B., B. C. Sykes, and R. E. M. Hedges
1995 Authenticating DNA extracted from ancient skel-
etal remains. Journal of Archaeological Science 22:
291299.
Roest Crollius, H., O. Jaillon, A. Bernot, C. Dasilva, L. Bou-
neaeu, C. Fischer, P. Wincher, P. Bottier, F. Quetier,
W. Saurin, and J. Weissenbach
2000 Estimate of human gene number provided by
genome-wide analysis using tetraodon nigroviridis
DNA sequence. Human Genetics 25:235238.
Rogan, P., and J. J. Salvo
1990 Molecular genetics of pre-Columbian South
American mummies. UCLA Symposia in Molecular
and Cellular Biology 122:223234.
Rogan, P. K., and J. J. Salvo
1990 Study of nucleic acids isolated from ancient remains.
Yearbook of Physical Anthropology 33:195214.
1994 High-fidelity amplification of ribosomal gene
sequences from South American mummies. In
Ancient DNA: Recovery and analysis of genetic material
from paleontological, archaeological, museum, medical,
and forensic specimens, edited by B. Herrmann and S.
Hummel, 182194. New York: Springer-Verlag.
Schneider, S., D. Roessli, and L. Excofer
2000 Arlequin 2.0: A software for population genetic data
analysis. Geneva, Switzerland, University of Geneva
Genetics and Biometry Laboratory.
Scholz, M., I. Giddings, and C. M. Pusch
1998 A polymerase chain reaction inhibitor of ancient hard
and soft tissue DNA extracts is determined as Human
Collagen Type I. Analytical Biochemistry 259:283286.
Schurr, T. G., S. W. Ballinger, and Y. Y. Gan
1990 Amerindian mitochondrial DNAs have rare Asian
mutations at high frequencies, suggesting they
derived from four primary maternal lineages.
American Journal of Human Genetics 46:613623.
Smith, D. G., R. S. Malhi, J. Eshleman, J. G. Lorenz, and F.
A. Kaestle
1999 Distribution of mtDNA haplogroup X among
Native North Americans. American Journal of Physi-
cal Anthropology 110:271284.
Smole, W. J.
1976 The Yanomama Indians: A cultural geography. Austin:
University of Texas Press.
Souza Aguiar, G. F., and W. A. Neves
1991 Postmarital residence and with-in sex genetic diver-
sity among the Urubu-Kaapor Indians, Brazilian
Amazon. Human Biology 63 (4): 467488.
Steadman, D. W.
1997 Population genetic analysis of regional and interre-
gional prehistoric gene ow in west-central Illinois.
Ph.D. dissertation, Department of Anthropology,
University of Chicago.
2001 Mississippians in motion? A population genetic
analysis of interregional gene ow in west-central
Illinois. American Journal of Physical Anthropology 114
(1): 6173.
Stone, A. C., and M. Stoneking
1993 Ancient DNA from a pre-Columbian Amerindian
population. American Journal of Physical Anthropology
92:463471.
1998 mtDNA analysis of a prehistoric Oneota population:
Implications for the peopling of the New World.
American Journal of Human Genetics 62:11531170.
Stoneking, M.
1995 Ancient DNA: How do you know when you have
it and what can you do with it? American Journal of
Human Genetics 57:12591262.
Sutter, R. C.
2000 Prehistoric genetic and culture change: A bioar-
chaeological search for pre-Inka altiplano colonies
in the coastal valleys of Moquegua, Peru, and Azapa,
Chile. Latin American Antiquity 11 (1): 4370.
152 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Thuesen, I., and J. Engberg
1990 Recovery and analysis of human genetic material
from mummied tissue and bone. Journal of Archaeo-
logical Science 17:679689.
Torroni, A., T. G. Schurr, M. F. Cabell, M. D. Brown, J. V.
Neel, M. Larsen, D. G. Smith, C. M. Vullo, and D.
C. Wallace
1993 Asian affinities and continental radiation of the
four founding Native American mtDNAs. American
Journal of Human Genetics 53:563590.
Tuross, N.
1994 The biochemistry of ancient DNA in bone. Experi-
mentia 50:530535.
Wallace, D. C., and A. Torroni
1992 American Indian prehistory as written in the
mitochondrial DNA: A review. Human Biology 64:
403416.
Ward, R. H.
1972 The genetic structure of a tribal population, the
Yanomama Indians. V. Comparisons of a series of
genetic networks. Annals of Human Genetics 36:
2143.
Ward, R. H., H. Gershowitz, M. Layrisse, and J. V. Neel
1975 The genetic structure of a tribal population, the
Yanomama Indians. XI. Gene frequencies for 10
blood groups and the ABH-Le secretor traits in the
Yanomama and their neighbors; the uniqueness of the
tribe. American Journal of Human Genetics 27:130.
Williams, S. R., L. A. Beck, and J. Longmire
1990 Human DNA recovery from ancient bone. American
Journal of Physical Anthropology 81:318.
Williams, S. R., N. A. Chagnon, and R. S. Spielman
2002 Nuclear and mitochondrial genetic variation in the
Yanomamo: A test case for ancient DNA studies of
prehistoric populations. American Journal of Physical
Anthropology 117 (3): 246259.
Williams, S. R., N. A. Doggett, and R. K. Moyzis
1992 Removing inhibitors using sepharose CL-6B.
Ancient DNA Newsletter 1:14.
Williams-Blangero, S., and J. Blangero
1990 Effects of population structure on within-group
variation in the Jirels of Nepal. Human Biology 62
(1): 131146.
Yang, D. Y., B. Eng, J. C. Dudar, S. R. Saunders, and J. S.
Waye
1997 Removal of PCR inhibitors using silica-based spin
columns: Application to ancient bones. Canadian
Society of Forensic Sciences Journal 30 (1): 15.
Zierdt, H., S. Hummel, and B. Herrmann
1996 Amplication of human short tandem repeats from
medieval teeth and bone samples. Human Biology 68
(2): 185199.
153
INTRODUCTION
Most papers in this volume concentrate on material
culture in addressing issues of identity and ethnicity.
This chapter, as well as those by Lozada and Buikstra,
Sutter, and Williams, will focus on the contribution that
analyses of human skeletal remains can add to the theme
of identity in past societies. Bioarchaeological data
contribute information that cannot be gained through
material culture alone and thus provide additional
insights toward archaeological inquiry. This is especially
important in the ancient Andes, where explanations of
cultural change often invoke population movement and
the convergence of peoples from diverse regions. In the
present study, I provide a bioarchaeological approach to
the investigation of Tiwanaku, one of the most exten-
sive and long-lived polities in the Andean region, with
two bioarchaeological foci: genetic relationships and
body modication.
The human body is one of the most powerful
and controversial means by which groups identify
themselves. In an ofcial statement of the American
Anthropological Association (1998), anthropologists
stress that it is biologically invalid to dene ethnic
or racial groups on the basis of discrete physical
differences and argue that human variation can gen-
erally be viewed as a continuum. Ethnicity and race
are essentially social constructs. However, highly vis-
ible features of an individual that cannot readily be
changed or controlled, for example physical features
such as skin or hair color, are often used as identiers
by others
1
(Rosenthal 1995; Royce 1982). Additionally,
humans often create distinct differences where they
do not exist biologically through an almost universal
propensity to modify and decorate bodies, embodying
cultural constructs even further. Such highly visible
cues generally are used to exchange information and
indicate group or ethnic boundaries (Hegmon 1992;
Jones and Hegmon 1991), and these modications
are sometimes present in the archaeological record.
Therefore, culturally constructed human bodies can
provide a powerful means of studying social dynamics
in ancient populations.
CHAP T E R 1 0
A Bioarchaeological Approach to
Tiwanaku Group Dynamics
DEBORAH E. BLOM
154 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Early European chroniclers described how ethnic
groups or tribes in the Andes distinguished them-
selves through distinct manners of emblems, dress, and
hairstyles (e.g., Casas 1967 [ca. 1550]:594595; Cieza
de Len 1984 [1553]:173; Garcilaso de la Vega 1966
[1609]:485; Vaca de Castro 1920 [1542]:18). Several of
these sources also document Andean groups that were
visibly distinguishing themselves from one another by
molding the heads of their infants into different shapes
(Casas 1892 [1561]; Cieza de Len 1984 [1553]:124;
Cobo 1990 [1653]:196197; Julien 1985:219; Murra
1980 [1956]:67). Further corroborating these claims,
distinct differences in cranial modication have been
noted in the archaeological record (Allison et al. 1981;
Arriaza 1988; Dembo and Imbelloni 1938; Gerszten
1993; Hoshower et al. 1995; Hrdlicka 1912; Lozada
1998; Munizaga 1964, 1976; OBrien and Sanzetenea
2002; Soto-Heim 1987; Stewart 1950; Weiss 1962,
1972). Permanent body modications such as this or
tattooing and scarring are specically those that tend to
denote group membership in general (Brain 1979; Ebin
1979; Isaacs 1975; Lyman and Douglass 1973; Rosen-
thal 1995; Royce 1982), and permanent modications
will be examined in this study.
The study of permanent style as expressed in cra-
nial modication complements that of more exible
stylistic variation in material culture, upon which
archaeological studies traditionally rely. Archaeolo-
gists cannot address ascription, the most critical aspect
of group membership, in which members recognize
themselves and others as belonging to distinct groups
holding different cultural values (Banks 1996; Barth
1969; Chapman 1993). However, these values are
frequently displayed indirectly in the cultural con-
tent or directly through overt signs or symbols. This
provides archaeologists an opportunity to distinguish
social groups through styles in material culture
(Aldenderfer and Stanish 1993; Conkey and Hastorf
1990; Jones 1997; Oakland Rodman 1992; Plog 1983;
Shennan 1989; Wiessner 1983).
Cranial modication differs from other expressions
of style, because individuals or groups may use cultural
material of another group or vary styles that they use
when changing residence, status, or afliation (Dietler
and Herbich 1989). Objects, ideas, and even ceramic
specialists (Hyslop 1993) can be moved independently
of populations, and archaeologists have long recognized
that inferring population movement or identity based
on material culture alone may not be entirely reliable
(Childe 1950:1; Hodder 1978; Trigger 1986:3947).
Consequently, cranial modication is more conservative
in reecting rapid changes in society compared to mate-
rial culture. Because it must be effected on the head of
an infant, it is an imposed symbol that reects the values
of the parents and society into which the child was born.
People may simultaneously have many identities, but
in cranial modication only one of these identities may
be expressed. Stylistic analyses of head form and how
these styles and others in material culture are used and
manipulated should prove especially fruitful in address-
ing social complexity and group afliation in the past.
Although it is recognized that group identity is
determined by ascription, a genetic component is
often present. Boundaries to biological reproduction are
often maintained through social means such as sexual
taboos and marriage rules or through geographic isola-
tion (Macbeth 1993). While social boundaries can be
crossed, interactions between ethnic groups are often
quite rigidly controlled in order to maintain these
boundaries (Barth 1969). On the other hand, in certain
cases, some individuals may be encouraged to marry
into other groups. For example, Kolata (1993) has
suggested that marriage alliances may have been made
between elites of different ethnic groups composing
Tiwanaku society. However, although smaller social
groups such as ayllus (or Andean descent groups) are
often exogamous, larger groups, such as ethnic groups,
are generally endogamous (Bastien 1978; Isbell 1977;
Rasnake 1988). Certainly, in the cases in which two
groups are geographically isolated, barriers to biologi-
cal relationships are further increased and even illicit
sexual relationships are not a factor.
In this study I use inherited, discrete variations in
teeth and bone to observe biological interactions between
groups. Inherited skeletal traits can be used to distinguish
between the movement of material culture or adoption of
styles (even cranial styles) and the movement of people.
Consequently, bioarchaeological data provide an impor-
tant complement to other studies in understanding social
dynamics in Tiwanaku, the focus of my study.
THE STUDY SAMPLE
With its capital centered on the shores of Lake Titicaca,
Tiwanaku ourished in the south-central Andes for
more than 600 years, roughly from 500 to 1100 AD.
Toward the end of this period, Tiwanaku-style mate-
rial culture could be found in regions of modern-day
Peru, Bolivia, Chile, and Argentina, making it one
of the most extensive pre-Inka polities. Extensive
B L O M 155
archaeological research over the last two decades has
signicantly increased our understanding of Tiwanaku
and the dynamics of sociopolitical complexity in the
highland demographic core (e.g., Albarracin-Jordn
1996a; Alconini Mjica 1995; Bermann 1994; Blom,
Janusek, and Buikstra 2003; Couture 2002, 2003;
Escalante 1992; Janusek 1994; Kolata 2003; Ponce
Sangins 1972; Rivera Casanovas 1994; Seddon 1994;
Stanish 1994; Vranich 1999).
Over the last ten to twenty years, signicant debate
has occurred over the exact nature of Tiwanaku
social complexity. Tiwanaku is now generally viewed
as encompassing various social groups, be they ayl-
lus (Albarracin-Jordn 1996b), moieties, or ethnic
groups (Janusek 1994; Kolata 1993; Ponce Sangins
1972). On one hand, some archaeologists emphasize
centralization, unication, and integration of diverse
groups within Tiwanaku (e.g., Kolata 1993). Others
maintain that Andean polities were always grounded
in local communities and argue that small-scale, local
political structure was sufficient for organizing the
construction of many of the impressive monumental
projects that others traditionally attribute to centralized
states (e.g., Albarracin-Jordn 1996b; see also Erickson
1993). More recently, archaeologists are discovering
that elements of both integration and local ties were
operating in Andean complex societies (e.g., Bermann
1994; Janusek 1994). In this view, while there is still
a shared Tiwanaku culture, local identity is displayed
within the larger context of a Tiwanaku identity.
Studies of Tiwanaku social dynamics have also been
carried out in the lowland regions in which Tiwanaku
material culture has been found. Most specically, for
the Moquegua Valley of southern Peru, Goldstein
(1989, 1993, 2000b) has proposed that Tiwanaku estab-
lished a diaspora community or provincial center, where
immigrants from the altiplano lived while maintaining
their altiplano identity (see also Blom et al. 1998;
Cohen, Bandy, and Goldstein 1995; Mujica, Rivera, and
Lynch 1983). However, Goldstein and his colleagues
acknowledge that the Tiwanaku population in the
Moquegua Valley could have been an autonomous
group who appeared culturally as Tiwanaku because
they were part of a Tiwanaku Interaction Sphere
(Moseley et al. 1991:123). This is consistent with oth-
ers who argue that Tiwanaku iconography and art may
have spread by llama caravans moving between the
altiplano and lowlands (Browman 1978, 1984; Kolata
1993; Lynch 1983). Therefore, the mere possession of
these artifacts might not indicate population movement
and shifts in identity. The present study builds upon this
work to address issues of identity within the Tiwanaku
realm. In this study, I use human skeletal samples from
the Tiwanaku heartland in the Bolivian altiplano and
from the distant settlements in the middle Moquegua
Valley (gure 1).
Arequipa
Puno
Moquegua
Azapa Valley
Osmore
Drainage
Ilo
Lake Titicaca
Lukurmata
Tiwanaku
0 300 km
N
VENEZUELA
BRAZIL
COLOMBIA
ECUADOR
PERU
BOLIVIA
CHILE
Key:
Modern settlements
Archaeological sites
Chen Chen
Figure 1. The study area.
156 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
The Moquegua sample consists of 436 individuals
and is primarily derived from the large site of Chen
Chen and secondarily from smaller sites in the valley
(table 1; gure 2). These collections are associated with
what was formerly considered the Chen Chen phase
(ca. AD 5001000; see Owen and Goldstein 2001)
and the earlier Huaracane (Formative) phase (ca. 385
BCca. AD 340 per Goldstein 2000a). The remains
from Moquegua are compared to two major areas in the
Lake Titicaca basin region: the Tiwanaku and Katari
valleys. The Tiwanaku Valley sample is dominated by
the large, urban site of Tiwanaku, which served as a
dense settlement of ~20,00040,000 inhabitants (Kolata
1993; Parsons 1968). The Katari Valley houses Lukur-
mata, a Tiwanaku regional center that seems to have
retained its own local tradition while becoming part of
Tiwanaku society (Bermann 1994; Janusek 1994), as
well as some smaller sites with these characteristics in
the nearby Pampa Koani, surrounding the site of Kirawi
(CK65) (gure 3). The samples from these valleys are
smaller, with 105 and 106 crania, respectively. Using the
most recent chronology for the Titicaca basin (Janusek
2003), the samples can be broken into Late Formative
and Tiwanaku period contexts from the Katari and
Moquegua valleys, and only Tiwanaku period contexts
are available from the Tiwanaku Valley (table 2). The
cranial modication sample consists of a smaller sample
of relatively complete crania, and the results will be
broken down into more precise time periods when pos-
sible. The biological distance sample includes individuals
from Chen Chen, the Moquegua Huaracane sites, and all
Tiwanaku and Katari valley sites mentioned.
Table 1. The sample used in the study.
Moquegua Valley Tiwanaku Sites: 406
Chen Chen (M1) [N=394]
Cancha de Yacango (M1666) [N=6]
Rio Muerto (M43) [N=4]
Cerro Trapiche (M7) [N=2]
Moquegua Valley Huaracane Sites: 30
Pampa Huaracane (M29, M30) [N=11]
Omo (M10) Y [N=19]
M162 [N=1]
Katari Basin Sites: 106
Lukurmata [N=78]
CK65 [N=15]
CK70 [N=9]
CK104 [N=2]
CK152 [N=2]
Tiwanaku Valley Sites: 105
Tiwanaku [N=95]
Tiwanaku Lower Valley:
Obsidiana [N=1]
Pukara [N=4]
Guaqui [N=2]
Iwawe [N=1]
Tiwanaku Middle Valley:
TMV101 [N=7]
TMV228 [N=3]
TMV332 [N=1]
TMV558 [N=1]
Total: 657
Figure 2. Osmore Drainage with detail of the Moquegua Valley sites mentioned.
B L O M 157
METHODS
Cranial Modication Methods
The cranial modication typology used in this study
is designed for viewing cranial modification from
a regional perspective. It was developed by Lozada
and myself through the study of cranial modication
in a large (N=255) series from the three coastal sites
(Chiribaya Alta, San Gernimo, Algodonal) and one
middle valley site (Yaral) from Moqueguas Chiribaya
and Tumilaca phases (Lozada 1998; Lozada, Blom,
and Buikstra 1997; see also gure 2). This system was
elaborated to extend to the Moquegua middle valley
and altiplano sites for this study. Because this typology
was designed to examine broad regional differences and
cranial modication as a marker of identity, externally
visible skull shape was considered in addition to the
more technologically based modication apparati. A
typology focused on technology that was developed
by Hoshower et al. (1995) through the study of thirty-
three skulls from cemeteries at Omo M10 and M11
2
was
also used for some comparisons within Moquegua. This
typology focused on the specic modication appara-
tus used, with the goal of discerning variation within
smaller cultural groups. After rst examining the same
Figure 3. Tiwanaku heartland. Source: after Kolata 1986: gure 2.
Table 2. Titicaca basin chronology (Janusek 2003).
Period Phase Time
Tiwanaku
Period
Late Tiwanaku V AD 1000 1100
Early Tiwanaku V AD 800 1000
Late Tiwanaku IV AD 600 800
Early Tiwanaku IV AD 500 600
Formative
Period
Tiwanaku I/III (Late Formative) 200 BC AD 500
Late Chiripa 800 BC 100 BC
Middle Chiripa 1000 BC 800 BC
Early Chiripa 1500 BC 1000 BC
158 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
crania studied by Hoshower and Buikstra, the typology
was applied to the Chen Chen sample (Blom, Yeatts,
and Buikstra 1995; Yeatts 1994). The typology for the
present study and that of Hoshower et al. overlap in a
statistically signicant manner (2 = 216.747, df=12,
p < 0.001, N = 131).
For the typology used here, the skulls were laid out
and grouped according to skull shape. All skulls were
mixed before classication in order to prevent biases
based on knowledge of the contextual information
by site, associated ceramic styles, or cemeteries. The
skulls were then sorted into main types, and then spe-
cics were recorded using the coding schema outlined
in table 3, which is a modied version of that found
in Buikstra and Ubelaker (1994). The procedure was
repeated twice to minimize error. If the crania were
too fragmented for the modication type to be deter-
mined, they were recorded as unmodied or modied
when possible.
Biological Distance Methods
Selection of Traits and Scoring
Although several methods of scoring inherited non-
metric traits have been adopted over the last century,
attempts to standardize data collection in bioarchaeology
in the last decade have resulted in general guidelines for
biological distance studies (e.g., Buikstra and Ubelaker
1994; Turner, Nichol, and Scott 1991). Most studies of
non-metric traits focus on cranial, postcranial, or dental
traits in isolation. However, using a large number of
traits from various parts of the body will ameliorate
any potential effects of local environment on the results
(Howe and Parsons 1967; Sofaer, Smith, and Kaye 1986).
For this reason and because some of the collections are
from geographic areas in which skeletal preservation is
not ideal, utilizing traits from the whole skeleton was
preferable in the present study.
The list of non-metric traits recorded for this study
comprises ninety-four cranial, postcranial, and dental
traits (appendix 1). These traits were selected
from bioarchaeological standards published
by Buikstra and Ubelaker (1994). The skel-
etal traits were compiled from numerous
sources (e.g., Finnegan 1978; Hauser and
De-Stefano 1989; Saunders 1978; Turner,
Nichol, and Scott 1991; Winder 1981),
while the dental traits compose a subset of
those described by Turner and colleagues
(1991). The rationale for selecting the traits
was based on ease of scoring and minimal
sexual dimorphism, as well as minimal intra-
and interobserver error. In order to further
minimize error, skeletal traits were scored
using the illustrations and descriptions
published in Buikstra and Ubelaker (1994)
and Finnegan (1978). Dental traits were
scored utilizing the procedures described
by Turner, Nichol, and Scott (1991) and the
graded dental casts supplied by the Arizona
State University Dental Laboratory.
Preparing the Data for Population
Comparisons
The possible methods of coding and analy-
sis for studies using non-metric traits vary
widely depending on the researcher and the
project. The various considerations are out-
lined here. See the appendix for the list of
traits recorded in the study and used in the
nal analysis. Traits were originally recorded
Table 3. Cranial Deformation Coding.
POSTERIOR DEFORMATION:
Number of Posterior Pads:
1=None
2=1 pad (midline)
3=2 pads (lateral to midline) (Note if
asymmetrical)
4=Band
Posterior Pad Location:
1=Absent
2=Lambda
3=Planum occipitale
4=Squamous portion of occipital
Plane of Pressure in Relation to
Transverse Plane:
1=Absent
2=Perpendicular (90)
3=Acute (<90)
4=Obtuse (>90)
Posterior Pad Shape:
1=Absent
2=Circular or oval
3=Donut-shaped
4=Triangular
5=Rectangular
6=Irregular form
7=Band
8=Unknown
Note impression of bindings
Sagittal Depression:
1=Absent
2=Slight near lambda (slight)
3=Almost to coronal suture (moderate)
4=Continuous (marked)
Note lambdic elevation or depression or
sagittal elevation
ANTERIOR DEFORMATION:
Number of Anterior Pads:
1=None
2=2 pads (lateral to midline) (Note if
asymmetrical)
3=1 pad (midline)
4=No denite pad impression, likely band
5=Band
Note bregmatic elevations
Anterior Pad Location:
1=Absent
2=High near coronal
3=Above bosses
4=Inferior, near or below bosses
Anterior Pad Shape:
1=Absent
2=Circular or oval
3=Donut-shaped
4=Triangular
5=Rectangular
6=Irregular form
7=Band
Note impression of bindings
Degree of Anterior Depressions:
1=Absent (Band, also)
2=Faint
3=Marked
Post-Coronal Constriction:
1=Absent
2=Slight
3=Marked
Post-Coronal Constriction Continuous:
1=Absent
2=Not Continuous
3=Continuous
B L O M 159
through partial trait scoring, but were later dichoto-
mized to reduce error. The dental traits were recoded
as presence/absence following standard methods (see
Sutter 1997; Turner 1985, 1987). Cranial and postcra-
nial non-metrics were collapsed into presence/absence
based on a similar scheme. Details are provided in the
appendix. Two traits were removed because they did not
vary overall (bregmatic bone and supratrochlear spur).
Many of the traits used in this study are bilateral.
Although it has been suggested that asymmetry is
genetically based (Ossenberg 1981), studies indicate
that asymmetry is caused by developmental factors
(Korey 1970; McGrath, Cheverud, and Buikstra 1984;
Saunders, Popovich, and Thompson 1978; Winder
1981), and the use of both sides will provide inaccu-
rate results. Because the two sides of bilateral traits are
invariably correlated (Finnegan 1978), it is necessary
to combine the data so that each trait is represented
only once when performing biodistance analyses.
Two basic methods have been suggested for dealing
with bilateral traits and their occasional asymmetry:
side-count methods, which involve adding all sides
as independent observations (e.g., Johnson and Lovell
1994; Prowse and Lovell 1996), and individual count
methods, in which each trait is only counted once per
individual (e.g., Buikstra 1976; Konigsberg 1990; Korey
1970; McGrath, Cheverud, and Buikstra 1984).
An individual count method was used in this study
because side-count methods artificially inflate the
sample size and, therefore, bias significance values
(Perizonius 1979). Although values can be calculated
using one, randomly selected side when both are avail-
able, the procedure used here follows the established
protocol for dental studies (Turner, Nichol, and Scott
1991; Turner and Scott 1977). This method assumes
that one genotype exists for bilateral traits, and, in
the case of asymmetry, the maximal expression of the
trait more closely approximates the underlying geno-
type. Therefore, only the maximal expression is used
for subsequent analyses. Unfortunately in the case of
only one side being available, it must be accepted as
the maximum expression, so a slight bias is introduced
in poorly preserved samples. However, since signi-
cant asymmetry is generally uncommon (Berry 1979;
Cosseddu, Floris, and Vona 1979), this is not expected
to bias the nal analyses.
Not all traits are used in the nal analyses because
biodistance analyses can be biased by low-frequency
traits or exceedingly small samples (Cavalli-Sforza,
Menozzi, and Piazza 1994). Traits were removed if:
1) the overall frequency was 0 or 100% in the total
sample; 2) the trait did not vary in at least two of the
samples; 3) less than ten observations were possible for
more than 25% of the samples; or 4) less than three
observations were possible for any of the samples.
Twenty-eight skeletal traits were removed because
they violated the necessary conditions for frequency
and sample size outlined above.
Clearly, some trait expression is age dependent in
children (Buikstra 1976). Therefore, the present study is
restricted to postadolescents. In addition, skeletal traits
that were signicantly different (p < 0.001) for the age
groups (postadolescent (1220), young adult (2034),
middle adult (3549), and old adult (50+)) were removed
from further analysis. The phi coefcient, which pro-
duces identical results as the Pearsons chi-square in
this sample, was used to determine signicance. Because
traits on the permanent dentition are not affected by
age, dental traits will be used for all individuals. It is
expected that the inclusion of dental traits will help to
ameliorate the effects of any small associations with age
in the skeletal traits. One skeletal trait (Allens fossa) was
removed, because it was highly correlated with age (phi
= 0.414, df = 3, p < 0.001).
In spite of initial reports of minimal linkage (Berry
and Berry 1967), several studies in humans have found
that some non-metric traits are associated with sex in
archaeological samples (Corruccini 1974; Saunders
1978). Two traits were removed because they were
highly sex dependent: rocker mandible (phi = -0.241,
df = 1, p < 0.001) and mandibular torus (phi = 0.151, df
= 1, p < 0.022).
Because the variables are not independent of one
another, using associated traits violates the assumptions
of many statistical procedures. In the present study,
intertrait associations were tested between traits in
similar body areas. In the case of highly signicant asso-
ciations (phi coefcient p < 0.001), hyperostotic traits
were chosen over other traits, which are less heritable
(Cheverud and Buikstra 1982). Dental traits are gener-
ally not intercorrelated, except the same trait within a
tooth type (i.e., Cusp 7 on the rst, second, and third
molars) (Sutter 1997). To avoid this intercorrelation,
only one tooth was used for each trait. Because sample
size is a concern, the best tooth was determined based
on the largest available sample. Of the traits that were
not removed due to previous considerations, two were
intercorrelated (p < 0.01): divided hypoglossal canal
within canal and divided hypoglossal canal-internal.
The within-canal variant was removed.
160 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Intra- and interobserver error must also be consid-
ered in studies of non-metric traits. Because intra- and
interobserver error is high for partial trait manifesta-
tions, the error is generally reduced by dichotomizing
the data into presence/absence observations and through
the use of the ASU dental casts (Molto 1983; Nichol
and Turner 1986; Saunders 1978, 1989). In order to
reduce error further, an initial learning period was
allowed in which the traits were observed on several
skeletons. The entire Chen Chen series was re-scored
after the initial scoring, because it was felt that it could
be more reliably recorded in this manner.
Intra- and interobserver error studies were carried
out on thirty individuals at Field Museum of Natural
History (Chicago, Illinois) and during the course of the
data collection. Interobserver error is not an issue in the
sample presented here, but was used for comparisons
presented elsewhere (Lozada 1998; Lozada et al. 1997).
The phi coefcient was used to evaluate the dichoto-
mized (presence/absence) data (Zar 1984:321323,
cited in Buikstra and Ubelaker 1994), and only one
remaining trait (Poiriers facet) had signicantly high
intraobserver error.
In contrast to the results of other researchers (e.g.,
Rothhammer et al. 1984; Sutter 1997), traits that did
not vary among the groups studied were not eliminated.
In a large set of traits, some differences between samples
will likely appear at random. Therefore, selecting such
traits for analysis might create artifactual differences
between samples.
Finally, some traits are affected by cranial modica-
tion, specically wormian bones in the cranial sutures
(Dorsey 1897; Guilln 1991; Konigsberg, Kohn, and
Cheverud 1993; Pucciarelli 1975). However, after all
traits that did not meet the requirements for use were
removed, no wormian bone traits remained.
Biodistance Analyses
Various means of determining biodistance can be used.
In the present study, a strictly model-free approach is
employed, and biodistance between each pair of sites
was determined using the mean difference of the trait
frequencies:
n
D
n
i
i i
f f
=
=
1
2 1
Where D is biodistance, f
1i
and f
2i
are frequencies
of the ith trait for populations 1 and 2, and n is the
number of traits. To calculate the probability that each
distance was obtained by chance in samples drawn
from a homogenous population (that the distance was
actually smaller), a bootstrap method was used. In this
method, the data set was subjected to a series of 1,000
randomizations in which each variable was indepen-
dently sorted according to random numbers. This has
the effect of randomizing the assignment of values to
individuals while maintaining the probability distribu-
tion of trait values for the whole sample. This simulates
the null hypothesis that the samples are drawn from the
same population.
Missing values are present in the data when the
trait was not observable for a particular individual
because the element was not present or was obscured
by soft tissue, pathological processes, or other artifacts.
These missing values were excluded from the frequency
calculations after randomization. Therefore, the fre-
quency of a trait was calculated as the number of times
the trait was scored as present divided by the sum of
present plus absent. In cases where all individuals
of a group had unobservable values for any trait, that
trait was eliminated for all comparisons with that group.
This applies both to the calculation of the actual dis-
tance and to the distances generated at each iteration by
the bootstrap. This procedure prevents the occurrence
of biasing results from fragmentary data sets where
many individuals have missing values. This test is,
therefore, sensitive to missing values only in that miss-
ing values will tend to inate the randomized variance
of within-sample trait frequencies and make signicant
p-values harder to obtain. This procedure was selected
in order to make results comparable to those of Lozada
(1998), who also used this method. Groups of less than
ten individuals were not used for analysis because very
small samples will not give useful results.
RESULTS
The Cranial Modication Types
The cranial modication typology resulted in the
following categories (see also gure 4):
Modication Absent
Cranial modication was classied as absent when
no apparent alteration in shape by a modifying device
was observed. Because this study focuses on intentional
modication, crania that had evidence of uninten-
tional modication resulting from the infant lying on
B L O M 161
a hard surface were also considered unmodied. The
skulls that were unintentionally modied present a
slight, acutely angled attening at lambda with no con-
current modication of the frontal. Overall, crania in
the absent category do not evince any obvious mark-
ing which could be attributed to modifying apparati.
Fronto-Occipital or Tabular Oblique
The crania corresponding to the fronto-occipital or
tabular oblique type of modication exhibit anterior-
posterior compression of the frontal and occipital
bones, which results in a relative shortening of the
parietal chord and lengthening of the occipital chord.
In extreme cases, the parietals expand laterally, result-
ing in highly prominent parietal bossing. This style was
effected by using tablets or rigid pads bound by thin
straps. Depending on the placement of the tablets or
pads, their relative exibility, and the angle at which
the pressure was applied, a variety of slightly different
cranial forms were produced, as seen in gure 4.
Circumferential or Annular
The annular oblique type is considerably different from
the fronto-occipital types of cranial modication and
would have produced a visible difference in head shape.
These crania were characterized by an elongated, tubu-
lar vault, produced by circumferential compression from
bands that encircled the frontal, temporal, parietal (below
the temporal line), and occipital bones. The modiers
used for this technique were probably constructed of
textiles and pads, or even thin exible pieces of wood.
This modification type for the present study differs
Figure 4. Cranial modication styles.
162 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
considerably from the annular type that we have previ-
ously reported for Chiribaya period skulls (Lozada 1998;
Lozada, Blom, and Buikstra 1997). On the Chiribaya
individuals, the bands were likely of an approximately
ten-centimeter width, while those discussed here were
thinner. Therefore, the skulls in the present study do not
have the loaf shape of those from Chiribaya contexts.
Circumferential pressure still represents the principal
modifying force in skulls belonging to this category.
Only one skull in the study did not t within the
typology outlined because the plane of modication in
the occipital region was distinctly perpendicular (tabu-
lar erect). The three main categories outlined (absent
(round), fronto-occipital (squashed) and annular
(tapered)) coincide with Aymara terms regarding
head form (Bertonio 1984 [1612]; table 4), and the
two modication styles are similar to those dened by
Anton (1989) (gure 5).
Table 4. Aymara terms regarding heads and head forms (italics indicates dictionary heading, whether in Spanish or Aymara) (Bertonio 1984 [1612]).
Aymara Spanish English
Ppekena, Cchuncchu, Chhikhana Cabeza head
Sayttu, Sucuya, Cchocata, Sau, Chunta,
Kistuna, Ppekena (e.g., Chunta
ppekenani)
Cabeza ahusada tapered head
Molloko, Moko, Mati ppekena (Mocona) Cabeza redonda round head
Phekena saytu De cabeza ahusada, cabeza sauu of tapered head
Phekena suticaa (or pallalla) De cabeza aplastada of squashed head
Sayttu Ppekenani De cabeza ahusada of tapered head
Sucuya Ppekenani De cabeza ahusada of tapered head
Cchacata ppekenani Same?
Cchocatha Ppekeani, o atar la cabeza con una trancadera, o por
enfermedad, o porque ahusndose la cabeza de los nios y nias
encajen mejor el sombrero, o capirote
tie or bind the head with a binding, or from
sickness, or because they taper the heads of
the children they t into the caps better
Cchocana La trancadera as the apparatus used
Cchocatha ppekenani De cabeza ahusada of tapered head
Figure 5. Superior view of crania (A. fronto-occipital, B. annular). Source: Anton 1989
B L O M 163
Upon examination of the cranial modication styles
from these samples, it is immediately apparent that
modication was practiced on the majority of the popu-
lation, since only 17% of the skulls (N=63/379) were
unmodied. The likelihood of an individual having an
unmodied head is not statistically different between
the valleys sampled (
2
= 0.48, df = 2, p = 0.976, N =
379). Neither modication in general (
2
= 0.001, df =
1, p = 0.976, N = 248) nor specic modication styles
(
2
= 3.174, df = 4, p = 0.529, N = 201) were associ-
ated with the sex of the individuals (see Blom 1999 for
more details). When the cranial modication styles
within Tiwanaku, Lukurmata, and Chen Chen and
their surrounding sites are analyzed, competing and
complementary patterns are visible.
The Moquegua Valley sample consists of two main
temporal contexts, one associated with the Tiwanaku
period (N=299) and the other with the earlier, For-
mative, Huaracane phase (N=5) (table 5). During the
Tiwanaku period, all individuals in Moquegua with
cranial modication were using fronto-occipital cranial
modication styles. This is also true of the Formative
period remains. Sixteen percent of the remains evinced
no head shape modication.
The Katari Valley sample also includes individuals
from the Formative (N=12) and Tiwanaku (N=19)
periods (table 6). In contrast to the patterns seen in
Moquegua, those buried at Lukurmata and the other
Table 5. Moquegua Valley Cranial Modication.
Modied Unmodied
Site(s) Time F/O Annular Total
N % N % N %
ChenChen Tiwanaku V 201 100% 45 16% 287
M1666 Tiwanaku 4 100% 2 33% 6
M43 Tiwanaku 3 100% 1 25% 4
M7 Tiwanaku 1 100% 2
Total Tiwanaku 209 100% 48 16% 299
Omo M10 Huaracane 4 100% 4
M162 Huaracane 1 100% 1
Total Huaracane 4 100% 1 20% 5
Total Moquegua 213 100% 49 16% 304
Table 6. Katari Valley Cranial Modication.
Modied Unmodied
Site(s) Time F/O Annular Total
N % N % N %
Lukurmata Tiwanaku I/III 6 100% 7
CK-65 Tiwanaku III 3 100% 4
CK-152 Late Chiripa 1 100% 1
Total Formative 10 100% 12
CK-152 Tiw III-Early IV 1 100% 1
CK-104.2 Tiwanaku IV 1 100% 1
Lukurmata Tiwanaku IV 3 100% 4 44% 9
CK-65 Tiwanaku V 3 100% 1 25% 4
CK-70 Tiwanaku V 1 100% 1
Lukurmata Tiwanaku V 1 50% 1 50% 3
Total Tiwanaku 1 10% 9 90% 6 33% 19
Total Katari Valley 1 5% 19 95% 6 19% 31
164 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Katari Valley sites were displaying predominately annu-
lar modication. Only one individual from a Tiwanaku
V context at Lukurmata was using a fronto-occipital
modication style. Nineteen percent of the individuals
overall had no cranial modication.
In contrast to the differences seen between the
Katari and Moquegua valleys, within the Tiwanaku
Valley and capital both tapered and fronto-occipitally
attened skulls were found (table 7). Roughly half of the
modied crania were found to be of each major style,
with 55% modied into the fronto-occipital form and
45% presenting the annular style. Fourteen percent of
the crania were unmodied.
When comparing the two modication types by area
with chi-square analyses, signicant differences in the
three cranial styles (absent, fronto-occipital, and annular)
were observed between highland (altiplano) and lowland
(Moquegua) populations (
2
= 158.455, df = 1, p < 0.001,
N = 262). The differences between the Moquegua, Katari,
and Tiwanaku valleys are also signicant (
2
= 182.724,
df = 2, p < 0.001, N = 262), as are those between the
major spatial/temporal groups of Moquegua/Formative,
Moquegua/Tiwanaku, Katari/Formative, Katari/
Tiwanaku, and Tiwanaku/Tiwanaku (
2
= 186.712, df =
4, p < 0.001, N = 262). Although I will not present the
extensive comparisons here (see Blom 1999, n.d.), no intra-
regional or intrasite patterns were seen in this sample.
Biodistance
Biological relationships can also add to the inter-
pretation. In order to fully understand the distance
measurements between the broad temporal/spatial
groups outlined in this chapter, it is helpful to look
at those between spatially distinct areas within the
cemetery of Chen Chen (table 8). Since it is largely
irrelevant for the present purposes, I do not include
the detailed map of these sectors; it can be found in
Blom (1999). The data indicate small distance measure-
ments that are statistically insignicant. This lack of
signicance indicates that the actual distances in the
populations are likely to have been smaller than those
seen in the study sample. Overall these data suggest that
the Chen Chen burial population was relatively homo-
geneous, an observation that is consistent with the data
on cranial modication for Moquegua. Furthermore,
these comparisons provide a general indication of what
biodistances can be considered small.
When the biological distance between the broad
spatial/temporal groups is analyzed, a pattern emerges
(table 9). As seen in the large p-values, comparisons
with the Katari Valley Formative sample do not pro-
duce signicant results (i.e., the distance measurements
could have actually been smaller in the populations).
This could, in part, be due to small sample sizes and
fragmented remains. Nevertheless, the distance mea-
surement between the Formative period Katari and
Moquegua valley samples is nearly signicant at the 0.05
level (p = 0.057; meaning that the distance is not likely
to be smaller), and the measurement is larger than that
seen elsewhere. This indicates that the largest biological
distance is between the Moquegua and Katari valleys
during the Formative period. The Katari/Tiwanaku
Table 7. Tiwanaku Valley Cranial Modication.
Modied Unmodied
Site(s) Time F/O Annular Total
N % N % N %
Tiwanaku Tiwanaku IV 1 33% 3
Tiwanaku Late Tiw IV 3 100% 1 25% 4
Tiwanaku Late IV-Early V 1 100% 1 33% 3
Tiwanaku Tiw IV-V 8 57% 6 43% 3 16% 19
Tiwanaku Tiwanaku V 4 67% 2 33% 6
Lower Tiwanaku Valley
a
Tiw IV-V 1 100% 1
Lower Tiwanaku Valley Tiwanaku V 2
Middle Tiwanaku Valley
b
Tiwanaku V 4 100% 6
Total Tiwanaku Period 16 55% 13 45% 6 14% 44
Total Tiwanaku Valley 16 55% 13 45% 6 14% 44
a
Pukara, Iwawe, Guaqui, Obsidiana
b
MV228, MV332, and Tilata
B L O M 165
sample also has a relatively large and quite signicant
distance from the Moquegua/Formative sample. In con-
trast, signicantly large distances are not found between
the Katari/Tiwanaku sample and the other samples. As
always, we are slightly limited by the lack of a Tiwanaku
Formative period sample.
Surprisingly, the distance between the Tiwan-
aku Valley (Tiwanaku/Tiwanaku) sample and the
Moquegua/Tiwanaku (Chen Chen) sample is as small
as that between samples within the Chen Chen site, as
seen previously in table 8. Other values which are not
signicant and are likely to be smaller than that between
Chen Chen and the Tiwanaku Valley sample are com-
parisons between the Katari/Tiwanaku sample and those
from Moquegua/Tiwanaku and Tiwanaku/Tiwanaku.
Therefore, the three Tiwanaku period groups of Chen
Chen, the Katari basin, and Tiwanaku are most closely
linked, with small biological distances. The most distant
measurements are between the Huaracane sample and
the Katari Valley samples. While the Huaracane sam-
ples closest biological distance is with Chen Chen, the
Chen Chen sample is closer to the altiplano Tiwanaku
period samples than it is to the Huaracane individuals.
Therefore, this would indicate more gene ow between
Chen Chen and Tiwanaku than between Chen Chen
and earlier Moquegua populations.
In summary, analyses between broad area/time
period groups in the entire region indicate: 1) the
largest distances between the Moquegua Formative
(Huaracane) and the Katari Valley Formative and
Tiwanaku period samples; 2) the smallest distances
between the three Tiwanaku period samples (Katari
Valley, Tiwanaku Valley, and Moquegua Valley); and
3) smaller distances between Moquegua and altiplano
Tiwanaku period samples than between the Moquegua
Valley Tiwanaku and Formative period samples.
DISCUSSION
Before assuming that cranial modication is linked to
ethnicity in Tiwanaku society, other possibilities must
be explored. Several lines of evidence can be considered
in explaining the results observed, and other possibili-
ties include status, gender, economic specialization, and
highland/lowland residence, most of which would not
necessarily exclude ethnicity as a factor.
Although it has been suggested by some that cra-
nial modication was practiced by only high-status
Table 8. Distance measurements and p-values for Chen Chen broad mortuary
areas.
Hilltop Ridge South
Hilltop
N = 56
Ridge
N = 107
d = 0.098
p = 0.121
South
N = 52
d = 0.110
p = 0.195
d = 0.106
p = 0.065
Slope
N = 120
d = 0.093
p = 0.201
d = 0.081
p = 0.088
d = 0.096
p = 0.185
Table 9. Distance measurements and p-values for context by time.
Context/Time
Number of Individuals
Katari /
Formative
Katari /
Tiwanaku
Tiwanaku/
Tiwanaku
Chen Chen
Katari/Formative
N = 12
Katari/Tiwanaku
N = 36
d = 0.139
p = 0.975
Tiwanaku/Tiwanaku
N = 72
d = 0.156
p = 0.840
d = 0.128
p = 0.134
Chen Chen/Tiwanaku
N = 335
d = 0.157
p = 0.771
d = 0.105
p = 0.253
d = 0.110
p = 0.000
Huaracane/Formative
N = 30
d = 0.244
p = 0.057
d = 0.208
p = 0.000
d = 0.161
p = 0.004
d = 0.146
p = 0.008
(statistically signicant values italicized)
166 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
individuals (e.g., dOrbigny 1944; Posnansky 1957), no
evidence exists to support the idea. The preponderance
of individuals makes the use of modication as a marker
of status differentiation unlikely. Likewise, the lack of
any signicant intrasite patterning in cranial modica-
tion styles also suggests that there is no link between
status and different types of modication. This is most
apparent in the core area of the Tiwanaku site, a sec-
tor that has been repeatedly recognized as inhabited
by elite individuals (e.g., Couture 2002; Janusek 1999;
Kolata 1993; Kolata and Ponce Sangins 1992). Previ-
ous studies have also shown no correlation between
cranial modication and tomb location, tomb type, or
grave offerings at Chen Chen (Blom 1999). DOrbigny
(1944:186) also claimed a correlation between gender
and altered head shape, but this suggestion is also not
supported with the current data. Therefore, neither
social status nor gender can explain cranial modica-
tion patterns seen here.
While observing the regional trend in cranial modi-
cation seen in this study, the oft-repeated truth
comes to mind that annular modication is present in
the highlands while tabular modication is found on the
coast (e.g., Bjrk and Bjrk 1964; Broca 1878; Squier
1973 [1877]: appendix B; Stewart 1950; Weiss 1962).
Based on the results from the Katari and Moquegua
valleys, it could seem possible that during the Middle
Horizon, the dichotomy between coastal (fronto-occipi-
tal) and highland (annular) styles holds. However, the
Tiwanaku Valley results denote that the distinction is
not nearly as clear as proposed, and individuals with
both annular and fronto-occipital variations are found
in the highlands during the Middle Horizon at least. In
addition, this pattern clearly does not hold for groups
in the immediate coastal area of the south Andes (e.g.,
Chiribaya and Chinchorro) from Archaic to Colonial
periods. In these populations, distinctive annular types
were also common, especially in contemporaneous
Chiribaya sites in the Moquegua region (Lozada 1998;
Lozada, Blom, and Buikstra 1997; see also Lozada and
Buikstra and Sutter, this volume).
Studies have also shown that in these Chiribaya sites
cranial modication types were linked to groups dif-
ferentiated by economic specialization (Lozada 1998).
Since areas of craft specialization have been noted
at Tiwanaku (Janusek 1999) and in Tiwanaku period
contexts in the Moquegua Valley (Goldstein 1993), this
is a potential explanation for the variation in cranial
modication in Tiwanaku society. However, because no
clear intrasite patterning was seen in the present study,
the hypothesis that cranial shape modication and eco-
nomic specialization are linked is not supported.
After ruling out other variables, ethnicity is a potential
explanation for the cranial modication patterns revealed
by this study. Certainly, the chroniclers described cer-
tain ethnic groups that distinguished themselves by head
forms (see Blom n.d. for details regarding this literature),
and these are precisely the types of body modication
that are often used as ethnic markers. However, the
hypothesis must be tested in each individual case, since
other groups can self-identify or be identied, such as
those based on gender, social class, and occupation, the
very groups that did not explain the cranial modication
data in this study. Jones (1997:84) denes ethnic groups
as culturally ascribed identity groups, which are based
on the expression of real or assumed shared culture and
common descent usually through the objectication of
cultural, linguistic, religious, historical, and/or physical
characteristics. Do the patterns that we see in Tiwanaku
society indicate ethnicity?
The patterning in cranial modication demonstrates
two regions with very distinct styles. In the Moquegua
Valley near the Pacic Coast, individuals used fron-
tal-occipital styles. In contrast, to the northeast of
Tiwanaku an annular type of modication predomi-
nated in the Katari Valley. Consequently, it is possible to
view the Tiwanaku realm as divided in two conceptually,
with Moquegua and fronto-occipital types to one side
and the Katari Valley and annular forms to the other.
Between these two regions, we nd the Tiwanaku Val-
ley with the capital, where individuals with both head
shapes were found. The border between east and west
is not a clearly demarcated line, but a dynamic bound-
ary where groups might converge. The patterning of
cranial modication styles also implies little or no per-
manent movement of individuals with modied skulls
between the east and west regions. On the surface, this
very much gives the impression of ethnic-group bound-
ary maintenance as described by Barth (1969). Since
ethnohistorical sources document it as such in other
areas, it can be reasonably proposed that cranial shape
modication was a stereotyped, overt sign of ethnic
ascription and differentiation key in the maintenance
of boundaries in Tiwanaku society. Studies of this Tiwa-
naku material culture can add to the interpretation of
these bioarchaeological data.
Various levels of ethnicity have been proposed
by archaeologists identifying spatial patterning in the
material culture throughout the Tiwanaku realm. Gold-
stein (1989, 1993, 2000b) argues that distant Tiwanaku
B L O M 167
settlements in the Moquegua Valley were composed of
a single ethnic group from Tiwanaku, or a diaspora
community, in which all the inhabitants identied sty-
listically with Tiwanaku. Owen and Goldstein (Owen
and Goldstein 2001) have also recently reinterpreted
Tiwanaku-afliated sites in Moquegua, which were
once thought to be temporally distinct on the basis of
ceramic styles, as two intraregional and contemporane-
ous ethnic groups. Unfortunately, the data presented
here t into only one of these groups. In the altiplano,
Janusek (1999, this volume; see also Bermann 1994)
documents subtle variation between compounds at
Tiwanaku and between Tiwanaku and Lukurmata in
terms of serving vessel form and decoration, household
practices, and differential access to particular resources
and ceramic styles. He argues that regional and house-
hold ethnic-like groups were actively displaying their
local identity within the broader identity as a member of
Tiwanaku society. Therefore, in many ways a correla-
tion between variation in ceramic styles and head form
exists. However, this correlation is not exact, and devia-
tions from it provide us with additional information.
As Barth (1969) points out in his classic piece on eth-
nicity, not all cultural stuff is relevant in maintaining
ethnic boundaries, since personnel and items usually
ow across those boundaries. The use of cranial modi-
cation and biodistance analyses allow us to directly
observe social mobility throughout an individuals
life and the boundaries to that mobility in Tiwanaku
society. Although there is a vast change in material
culture between the Formative and Tiwanaku periods,
altered cranial shapes remained the same in the Katari
and Moquegua valleys. This would mean, if one takes
a primordial approach to ethnicity, that the boundar-
ies of the groups remained unchanged even though
other cultural elements were transformed, unless, of
course, the meaning of modification changed over
time (see Hensel 1996). Cranial modication does not
vary between structures in the site of Tiwanaku, even
though Janusek documents distinctions in material
culture. However, the differences he outlines between
Lukurmata and Tiwanaku can be seen in the dissimilar-
ity between the Katari and Tiwanaku valleys.
By focusing on boundaries, Barth discusses how iden-
tity is maintained when groups interact, deviating from
past assumptions that such groups become homogenized
over time. Based on migration studies from Western
Europe, a common assumption was that, in general, inte-
gration resulted in a loss of kinship bonds (du Toit 1975:
9; see also Childe 1950; Park 1950; and Weber 1958,
1979). However, there are many examples of people
migrating specically to areas where they have family,
and it is very common to retain kin ties in the homeland
after migration (du Toit 1975; Lewis 1965). Even in situ-
ations when no kin ties exist, ctive-kinship (compadrazgo
or ritual kinship) becomes important. Therefore, we have
no reason to expect that those who are physically or emo-
tionally integrated into a larger whole would not retain
their ties in their perceived homeland, even while cre-
ating new relationships. The likelihood of maintaining
homeland identity is further increased by the common
tendency to visit or return home, and kin may act as a
bridge between two communities (Anthony 1990; du
Toit 1975). This would be true in the case of migration to
or from Tiwanaku. People have a clear capacity to retain
many identities at one time, even those that may seem
incongruous to an outsider.
Data on inherited skeletal traits through biodistance
analyses add more information to the interpretation.
Biodistance analyses indicated an increase in gene ow
between regions from the Formative to Tiwanaku
periods. In the Formative period, populations from the
Katari and Moquegua valleys were quite distant biologi-
cally, appearing to have been generally isolated from
one another. During the Tiwanaku period, biological
distance measures show a drastic change in which bio-
distances between all regions become quite small, likely
indicating movement of people between these areas or
at least an increase in sexual reproduction between
individuals from these regions. Therefore, in contrast
to what was seen earlier, cranial modication seems to
have posed no real or strict barrier to sexual reproduc-
tion while Tiwanaku material culture styles were being
used by the larger region. In fact, Tiwanaku culture
in some ways encouraged the convergence of diverse
groups, at least temporarily if not permanently.
During the Tiwanaku period, diverse people from
a large region were using a common material culture
style. Tiwanaku style has been generally viewed as
emerging from the highland core and spreading
into the peripheral areas where regional variants (e.g.,
Cochabamba and Moquegua) were developed (Bennett
1936:402; Ponce Sangins 1972; Rydn 1957, 1959).
However, based on the present data, this pattern can
best be characterized as Tiwanaku drawing individuals
from outlying homogeneous areas into a heterogeneous
center, or boundary. This is consistent with recent work
on migration that points out that migration is very
rarely one-way (Anthony 1990). We can probably best
view population movement in and out of Tiwanaku as
168 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
cyclical. Yet, through rules of interaction, sociocultural
boundaries were maintained, as seen in data on cranial
modication styles and biological distance. Only in the
capital do we visibly see individuals of diverse groups
living and buried together. However, outside of the
boundary and capital, we see a strong sense of local,
perhaps ayllu, identity being displayed with homogene-
ity in culturally constructed head shape.
CONCLUSION
Using different forms of style in material culture,
archaeologists have addressed the issue of group
membership and diversity in Tiwanaku. The present
study illustrates the importance of a bioarchaeological
approach in addressing social complexity and group
dynamics in Tiwanaku society. This study also demon-
strates that boundaries can be seen in Tiwanaku society.
However, the complexity of these boundaries is only
apparent when studies include both the heartland and
peripheries. In this way, the Tiwanaku Valley during
the Tiwanaku period can be viewed as a distinct bor-
derland at which interactions between neighboring
regions occurred.
Through this analysis of permanent body modica-
tion and biological distance, two major regions in the
Tiwanaku realm are revealed, each of which maintained
a sense of a clear boundary while incorporated within
Tiwanaku society. The capital city of Tiwanaku is envi-
sioned as a cosmopolitan center, welcoming individuals
of the diverse neighboring regions. Tiwanaku expansion
is viewed as more than individuals from the highlands
spreading into the lowlands during the later Tiwanaku
period. These data indicate that individuals may have
moved to and fro, perhaps in areas in which they already
had some local ties.
It is hoped that the methods outlined here may
allow us to utilize a broader array of collections in
studying Andean social complexity. It is recognized that
nondomestic contexts, such as mortuary areas, are not
suitable independently for testing archaeological mod-
els (Stanish 1992). Mortuary contexts have been shown
to contain exotic material, and the models necessitate
detecting information about the local population. Using
cranial modication styles and inherited traits, we can
access a portion of the mortuary context that is funda-
mentally local, allowing mortuary remains, which make
up a large part of the collections housed in the Andes,
to be used for a wider range of inquiry.
ACKNOWLEDGEMENTS
The material presented here in part is derived from works
published in World Archaeology and Journal of Anthropo-
logical Archaeology. This research was funded by a grant
from the Wenner Gren Foundation for Anthropological
Research, with important additional support from Jane
Buikstra in the early days of the project. Many colleagues
kindly provided very helpful comments and encourage-
ment on various versions of this research, including
Buikstra, Nicole Couture, Benedikt Hallgrmsson,
Janusek, Linda Keng, and Alan Kolata. Nevertheless,
any errors are my own. Paul Goldstein, Bruce Owen, and
Bertha Vargas were generous in providing collections and
contextual information for the Moquegua Valley samples.
Likewise, Kolata and Javier Escalante of UNAR helpfully
allowed access to the Tiwanaku collection. An amazing
number of people helped during curation and analysis of
the Moquegua and Tiwanaku collections. They include
Don Genaro, Jahel Amaru, Rosala Choque Gonzales,
Etty Indriati, Liz Klarich, Carla Lee, Arminda Mamani,
Raul Menaut, Santiago Morales, Carrie Oehler, Agustn
Paty, Bonnie Podest, Henry Tantalen, Bill Taylor,
Elva Torres, Danilo Villamor, and Dale Yeatts. Benedikt
Hallgrmsson provided the statistical analyses for the
biodistance measurements used here and in Lozadas
work. Susan Anton generously granted permission to use
gure 5. Editorial aid by Cindy Longwell was essential
in completing this project. I am also grateful to many
others not listed by name who were also a vital part of
this entire process.
NOTES
1. Less visible differences may be used to differentiate
within groups, but highly visible traits are generally those
used by outsiders to identify group membership.
2. This sample consisted of twenty-four skulls from Tiwa-
naku V phase contexts from Omo M10; four Tumilaca phase
individuals from M10; four Huaracane phase skulls from M10;
and one Tumilaca phase cranium from Omo M11 (Hoshower
et al. 1995:5455).
REFERENCES
Albarracin-Jordn, J. V.
1996a Tiwanaku: Arqueologa regional y dinmica segmentara.
La Paz, Bolivia: Plural Editores.
1996b Tiwanaku settlement system: The integration of
nested hierarchies in the lower Tiwanaku Valley.
Latin American Antiquity 7 (3): 183210.
B L O M 169
Alconini Mjica, S.
1995 Rito, smbolo, e historia en la pirmide de Akapana,
Tiwanaku: Un anlisis de cermica ceremonial prehis-
pnica. La Paz, Bolivia: Editorial Accin.
Aldenderfer, M. S., and C. Stanish
1993 Domestic architecture, household archaeology,
and the past in the south-central Andes. In Domes-
tic architecture, ethnicity, and complementarity in the
south-central Andes, edited by M. S. Aldenderfer,
112. Iowa City: University of Iowa Press.
Allison, M. J., E. Gerszten, J. Munizaga, C. Santoro, and G.
Focacci
1981 La prctica de la deformacin craneana entre los pueb-
los andinos precolombinos. Chungar 7:238260.
American Anthropological Association
1998 AAA Statement on Race. Anthropology Newsletter
39 (6): 3.
Anthony, D. W.
1990 Migration in archaeology: The baby and the bath-
water. American Anthropologist 92 (3): 895914.
Anton, S. C.
1989 Intentional cranial vault deformation and induced
changes of the cranial base and face. American Jour-
nal of Physical Anthropology 79:253267.
Arriaza, B.
1988 Modelo bioarqueolgico para la bsqueda y acerca-
miento al individuo social. Chungar 21:932.
Banks, M.
1996 Ethnicity: Anthropological constructions. New York:
Routledge.
Barth, F.
1969 Ethnic groups and boundaries: The social organization of cul-
ture difference. Boston: Little, Brown and Company.
Bastien, J. W.
1978 Mountain of the condor: Metaphor and ritual in an
Andean ayllu. American Ethnological Society Mono-
graph 64. New York: West Publishing Company.
Bennett, W. C.
1936 Excavations in Bolivia. Anthropological Papers of the
American Museum of Natural History 35 (4). New
York.
Bermann, M. P.
1994 Lukurmata: Household archaeology in prehispanic
Bolivia. Princeton, NJ: Princeton University Press.
Berry, A. C., and R. J. Berry
1967 Epigenetic variation in the human cranium. Journal
of Anatomy 101:361379.
Berry, R. J.
1979 Section I. Genes and skeletons, ancient and modern.
Journal of Human Evolution 8:669677.
Bertonio, P. L.
1984 [1612] Vocabulario de la lengua Aymara. Cochabamba:
CERES/ IFEA/ MUSEF.
Bjrk, A., and L. Bjrk
1964 Articial deformation and craniofacial asymmetry
in ancient Peruvians. Journal of Dental Research 43:
353362.
Blom, D. E.
1999 Tiwanaku regional interaction and social identity:
A bioarchaeological approach. Ph.D. disserta-
tion, Department of Anthropology, University of
Chicago.
n.d. Embodying borders: Human body modication and
diversity in Tiwanaku society. Journal of Anthropo-
logical Archaeology.
Blom, D. E., B. Hallgrmsson, L. Keng, M. C. Lozada C.,
and J. E. Buikstra
1998 Tiwanaku colonization: Bioarchaeological impli-
cations for migration in the Moquegua Valley, Peru.
World Archaeology 30 (2): 238261.
Blom, D. E., J. W. Janusek, and J. E. Buikstra
2003 A re-evaluation of human remains from Tiwanaku.
In Tiwanaku and its hinterland: Archaeology and paleo-
ecology of an Andean civilization, vol. 2 of Urban and
rural archaeology, edited by A. L. Kolata, 435448.
Washington, D.C.: Smithsonian Institution Press.
Blom, D. E., M. C. Lozada C., B. Hallgrmsson, L. Keng,
and J. E. Buikstra.
1997 Biology and culture in the prehispanic south-central
Andes. Paper presented at the 62nd Annual Meeting
of the Society for American Archaeology, April 2,
1997. Nashville.
Blom, D. E., D. J. Yeatts, and J. E. Buikstra
1995 A bioanthropological approach to Tiwanaku-
Moquegua Valley interaction. Paper presented at
the 60th Annual Meeting of the Society for Ameri-
can Archaeology, May 4, 1995. Minneapolis.
Brain, R.
1979 The decorated body. New York: Harper and Row.
Broca, P.
1878 Sur des crnes et des objets dindustrie provenant des
fouilles de M. Ber Tiahuanaco (Prou). Bulletins de
la Societ dAnthropologie; Pars, 3a serie 1:230235.
Browman, D.
1978 Toward the development of the Tiahuanaco (Tiwa-
naku) state. In Advances in Andean archaeology, edited
by D. L. Browman, 327349. The Hague: Mouton
Publishers.
1984 Tiwanaku: Development of interzonal trade and
economic expansion in the Altiplano. In Social
170 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
and economic organization in the prehispanic Andes,
edited by D. L. Browman, R. L. Burger, and M.
A. Rivera, 117142. BAR International Series 194,
Proceedings of the 44th International Congress of
Americanists, Manchester, 1982. Oxford: British
Archaeological Reports.
Buikstra, J.
1976 Hopewell in the lower Illinois Valley: A regional approach
to the study of human biological variability and prehistoric
behavior. Scientic Papers 2. Evanston, IL: North-
western University Archaeological Program.
Buikstra, J., and D. Ubelaker, eds.
1994 Standards for data collection from human skeletal
remains: Proceedings of a seminar at the Field Museum
of Natural History. Arkansas Archaeological Survey
Research Series 44. Fayetteville.
Casas, B. de las
1892 [1561] De las antiguas gentes del Per. Madrid: Manuel
G. Hernandes.
1967 [ca. 1550] Apologtica historia sumaria, vol. 2 (compiled
by Edmundo OGorman). Mexico City: Universidad
Nacional Autnoma de Mxico, Instituto de Inves-
tigaciones Histricas.
Cavalli-Sforza, L. L., P. Menozzi, and A. Piazza
1994 The history and geography of human genes. Princeton,
NJ: Princeton University Press.
Chapman, M.
1993 Social and biological aspects of ethnicity. In Social
and biological aspects of ethnicity, edited by M. Chap-
man, 146. New York: Oxford University Press.
Cheverud, J. M., and J. E. Buikstra
1982 Quantitative genetics of skeletal non-metric traits in
the rhesus macaques on Cayo Santiago. III. Relative
heritability of skeletal non-metric and metric traits.
American Journal of Physical Anthropology 59:151155.
Childe, V. G.
1950 Prehistoric migrations in Europe. Oslo: H. Aschehoug
and Company (W. Nygaard).
Cieza de Len, P. de
1984 [1553] La crnica del Per. Monumenta Hispano-Indiana
V Centenario del Descumbrimiento de Amrica II.
Madrid: Consejo Superior de Investigaciones Cient-
cas, Instituto Gonzalo Fernndez de Oviedo.
Cobo, B.
1990 [1653] Inca religion and customs (selections from Historia
del Nuevo Mundo). Translated and edited by R.
Hamilton. Austin: University of Texas Press.
Cohen, A. B., M. Bandy, and P. Goldstein
1995 How old is that archipelago: The Huaracane tradition
and the antiquity of vertical control in the south Andes.
Paper presented at the 35th Annual Meeting of the
Institute for Andean Studies, 1995. Berkeley, CA.
Conkey, M. W., and C. A. Hastorf, eds.
1990 The uses of style in archaeology. Cambridge: Cam-
bridge University Press.
Corruccini, R. S.
1974 An examination of the meaning of cranial discrete
traits for human skeletal biological studies. American
Journal of Physical Anthropology 40:425446.
Cosseddu, G. G., G. Floris, and G. Vona
1979 Sex and side differences in the minor non-metrical cra-
nial variants. Journal of Human Evolution 8:685692.
Couture, N. C.
2002 The construction of power: Monumental space and
elite residence at Tiwanaku, Bolivia. Ph.D. disserta-
tion, Department of Anthropology, University of
Chicago.
2003 Ritual, monumentalism, and residence at Mollo
Kontu, Tiwanaku. In Tiwanaku and its hinterland:
Archaeology and paleoecology of an Andean civilization,
vol. 2 of Urban and rural archaeology, edited by A. L.
Kolata. 202225. Washington, D.C.: Smithsonian
Institution Press.
Dembo, A., and J. Imbelloni
1938 Deformaciones intencionales del cuerpo humano de
carcter tnico. Buenos Aires: Jose Anesi.
Dietler, M., and I. Herbich
1989 Tich Matek: The technology of Luo pottery produc-
tion and the denition of ceramic style. (Ceramic
Technology). World Archaeology 21 (1): 148165.
dOrbigny, A.
1944 El hombre Americano: Considerado en sus aspectos si-
olgicos y morales. Buenos Aires: Editorial Futuro.
Dorsey, C. A.
1897 Wormian bones in articially deformed Kwakiutl
crania. American Anthropology 10:169172.
du Toit, B. M.
1975 Introduction: Migration and population mobility.
In Migration and urbanization: Models and adaptive
strategies, edited by B. M. du Toit and H. I. Safa,
115. The Hague: Mouton Publishers.
Ebin, V.
1979 The body decorated. London: Thames and Hudson.
Erickson, C. L.
1993 The social organization of prehispanic raised eld
agriculture in the Lake Titicaca basin. In Economic
aspects of water management in the prehispanic New
World, edited by V. L. Scarborough and B. L. Isaac,
369426. Research in Economic Anthropology,
suppl. 7. Greenwich: JAI Press.
B L O M 171
Escalante, J.
1992 Excavaciones arqueolgicas en la Karaa. Internal
report to the Unidad Nacional de Arqueologa. La
Paz, Bolivia.
Finnegan, M.
1978 Non-metric variation of the infracranial skeleton.
Journal of Anatomy 125 (1): 2337.
Garcilaso de la Vega, S.
1966 [1609] Royal commentaries of the Incas and general history
of Peru. Austin: University of Texas Press.
Gerszten, P. C.
1993 An investigation into the practice of cranial deformation
among the pre-Columbian peoples of northern Chile.
International Journal of Osteoarchaeology 3 (2): 8798.
Goldstein, P. S.
1989 Omo, a Tiwanaku provincial center in Moquegua,
Peru. Ph.D. dissertation, Department of Anthropol-
ogy, University of Chicago,
1993 Tiwanaku temples and state expansion: A Tiwan-
aku sunken-court temple in Moquegua, Peru. Latin
American Antiquity 4 (1): 2247.
2000a Exotic goods and everyday chiefs: Long-distance
exchange and indigenous sociopolitical develop-
ment in the south-central Andes. Latin American
Antiquity 11 (4): 107131.
2000b Communities without borders: The vertical archi-
pelago, and diaspora communities in the southern
Andes. In The archaeology of communities: A New
World perspective, edited by J. Yaeger and M. Canuto,
182209. New York: Routledge Press.
Guilln, S.
1991 The Chinchorro adaptation to the environment of
the south-central Andes. Ph.D. dissertation, Depart-
ment of Anthropology, University of Michigan.
Hauser, G., and G. F. De-Stefano
1989 Epigenetic variants of the human skull. Stuttgart: Sch-
weizerbart.
Hegmon, M.
1992 Archaeological research on style. Annual Review of
Anthropology 21:517536.
Hensel, C.
1996 Telling our selves: Ethnicity and discourse in southwestern
Alaska. New York: Oxford University Press.
Hodder, I.
1978 Simple correlations between material culture and
society: A review. In The spatial organisation of culture,
edited by I. Hodder, 324. Pittsburgh: University
of Pittsburgh Press.
Hoshower, L. M., J. E. Buikstra, P. S. Goldstein, and A. D.
Webster
1995 Articial cranial deformation in the Omo M10 site:
A Tiwanaku complex from the Moquegua Valley,
Peru. Latin American Antiquity 6 (2): 145164.
Howe, W. L., and P. A. Parsons
1967 Genotype and environment in the determination
of minor skeletal variants and body weight in mice.
J. Embryol. Exp. Morph. 17:283292.
Hrdlicka, A.
1912 Artificial deformation of the human skull with
especial (sic) reference to America. In Actas del
XVII Congreso Internacional de Americanistas, Sesin
de Buenos Aires, 17-23 de Mayo de 1910, edited by
R. Lehmann-Nitsche (Secretario General del Con-
greso), 147149. Buenos Aires: Imprenta de Coni
Hermanos.
Hyslop, J.
1993 Factors inuencing the transmission and distribution
of Inka cultural materials throughout Tawantinsuyu.
In Latin American Horizons, edited by D. S. Rice,
337356. Washington, D.C.: Dumbarton Oaks
Research Library and Collection.
Isaacs, H. R.
1975 Idols of the tribe: Group identity and political change.
New York: Harper and Row.
Isbell, B. J.
1977 Those who love me: An analysis of Andean kinship
and reciprocity within a ritual context. In Andean
kinship and marriage, edited by R. Bolton and E.
Mayer, 81105. American Anthropological Asso-
ciation Special Publications 7. Washington, D.C.:
American Anthropological Association.
Janusek, J. W.
1994 State and local power in a prehispanic Andean
polity: Changing patterns of urban residence in
Tiwanaku and Lukurmata, Bolivia. Ph.D. disser-
tation, Department of Anthropology, University
of Chicago.
1999 Craft and local power: Embedded specialization in
Tiwanaku cities. Latin American Antiquity 10 (2):
107131.
2003 Vessels, time, and society: Toward a chronology of
ceramic style in the Tiwanaku heartland. In Tiwa-
naku and its hinterland: Archaeology and paleoecology
of an Andean civilization, vol. 2 of Urban and rural
archaeology, edited by A. L. Kolata, 3094. Wash-
ington, D.C.: Smithsonian Institution Press.
Johnson, A. I., and N. C. Lovell
1994 Biological differentiation at Predynastic Naqada,
Egypt: An analysis of dental morphological traits.
American Journal of Physical Anthropology 93:427433.
172 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Jones, K., and M. Hegmon
1991 The medium and the message: A survey of informa-
tion conveyed by material culture in middle range
societies. Submitted to paper presented at the
56th Annual Meeting of the Society for American
Archaeology, New Orleans.
Jones, S.
1997 The archaeology of ethnicity: Constructing identities in
the past and present. New York: Routledge.
Julien, C. J.
1985 Guano and resource control in sixteenth-century
Arequipa. In Andean ecology and civilization, edited
by S. Masuda, I. Shimada, and C. Morris, 185231.
Tokyo: University of Tokyo Press.
Kolata, A. L.
1986 The agricultural foundations of the Tiwanaku state:
A view from the heartlands. American Antiquity 51:
746-762.
1993 Tiwanaku: Portrait of an Andean civilization. Cam-
bridge: Blackwell.
Kolata, A. L., ed.
2003 Tiwanaku and its hinterland: Archaeology and paleo-
ecology of an Andean civilization, vol. 2 of Urban and
rural archaeology. Washington, D.C.: Smithsonian
Institution Press.
Kolata, A. L., and C. Ponce Sangins
1992 Tiwanaku: The city at the center. In The ancient
Americas: Art from sacred landscapes, edited by R. F.
Townsend, 317334. Chicago: The Art Institute.
Konigsberg, L. W.
1990 Temporal aspects of biological distance: Serial cor-
relation and trend in a prehistoric skeletal lineage.
American Journal of Physical Anthropology 82:4552.
Konigsberg, L. W., L. A. P. Kohn, and J. M. Cheverud
1993 Cranial deformation and non-metric trait variation.
American Journal of Physical Anthropology 90:3548.
Korey, K.
1970 Characteristics of the distribution of non-metric
variants. Masters thesis, Department of Anthro-
pology, University of Chicago.
Lewis, O.
1965 Urbanization without breakdown: A case study. In
Contemporary cultures and societies of Latin America:
A reader in the social anthropology of Middle and South
America and the Caribbean, edited by D. B. Heath
and R. N. Adams, 424437. New York: Random
House.
Lozada, M. C.
1998 The seoro of Chiribaya: A bioarchaeological
study in the Osmore Drainage of southern Peru.
Ph.D. dissertation, Department of Anthropology,
University of Chicago.
Lozada, M. C., D. E. Blom, and J. E. Buikstra
1997 La deformacin craneana articial: Una prctica
cultural entre los pobladores Chiribaya de Ilo,
sur del Per. Paper presented at Seminario sobre
Arqueologa de Ilo, Ponticia Universidad Catlica
del Per, August 24, 1997. Lima.
Lozada, M. C., B. Hallgrmsson, D. E. Blom, L. Blue, S.
Guillen, and J. E. Buikstra
1997 Patrones microevolucionarios y poblaciones bioar-
queolgicas en la costa del Osmore, sur del Per.
Paper presented at Seminario sobre Arqueologa
de Ilo, Ponticia Universidad Catlica del Per,
August 23, 1997. Lima.
Lyman, S. M., and W. A. Douglass
1973 Ethnicity: Strategies of collective and individual
impression management. Social Research 40:344365.
Lynch, T. F.
1983 Camelid pastoralism and the emergence of Tiwa-
naku civilization in the south-central Andes. World
Archaeology 15:114.
Macbeth, H.
1993 Ethnicity and human biology. In Social and biological
aspects of ethnicity, edited by M. Chapman, 4791.
New York: Oxford University Press.
McGrath, J. W., J. M. Cheverud, and J. E. Buikstra
1984 Genetic correlations between sides and heritabil-
ity of asymmetry for non-metric traits in rhesus
macaques on Cayo Santiago. American Journal of
Physical Anthropology 64:401411.
Molto, J. E.
1983 Biological relationships of southern Ontario Woodland
peoples: The evidence of discontinuous cranial morphol-
ogy. National Museums of Canada, Archaeological
Survey of Canada, Paper 117. Ottawa.
Moseley, M. E., R. A. Feldman, P. S. Goldstein, and L.
Watanabe
1991 Colonies and conquest: Tiahuanaco and Huari in
Moquegua. In Huari administrative structure: Pre-
historic monumental architecture and state government,
edited by W. H. Isbell and G. F. McEwan, 121140.
Washington, D.C.: Dumbarton Oaks.
Mujica, E., M. A. Rivera, and T. F. Lynch
1983 Proyecto de estudio sobre la complementariedad
econmica Tiwanaku en los valles occidentales del
centro-sur andino. Chungar 11:85109.
Munizaga, J. R.
1964 Deformacin ceflica intencional (anlisis algunas
poblaciones precolombinas en el norte de Chile).
B L O M 173
Antropologa, Revista del Centro de Estudios Antrop-
olgicos, Universidad de Chile 2 (2): 518.
1976 Intentional cranial deformation in the pre-Colum-
bian population of Ecuador. American Journal of
Physical Anthropology 45:687694.
Murra, J. V.
1980 [1956] The economic organization of the Inka state.
Research in Economic Anthropology, suppl. 1.
Greenwich, CN: JAI Press, Inc. Reprint of The
Economic Organization of the Inka State, 1956,
Ph.D. dissertation, University of Chicago.
Nichol, C. R., and C. G. Turner, II
1986 Intra- and interobserver error concordance in
classifying dental morphology. American Journal of
Physical Anthropology 69 (3): 299315.
Oakland Rodman, A.
1992 Textiles and ethnicity: Tiwanaku in San Pedro de
Atacama, North Chile. Latin American Antiquity 3
(4): 316340.
OBrien, T. G., and R. Sanzetenea
2002 Deformacin craneana articial: Un estudio de la coleccin
de Ibarra Grasso de Cochabamba. Boletn del INIAN
Museo. Serie Arqueolgica Boliviana 4 (26), Enero-
Febrero. Cochabamba, Bolivia: Museo Arqueolgico
de la Universidad Mayor de San Simn.
Ossenberg, N. S.
1976 Within and between race distances in popula-
tion studies based on discrete traits of the human
skull. American Journal of Physical Anthropology 45:
701716.
1981 An argument for the use of total side frequencies
of bilateral non-metric skeletal traits in population
distance analysis: The regression of symmetry on
incidence. American Journal of Physical Anthropology
54:471479.
Owen, Bruce, and Goldstein, Paul S.
2001 Tiwanaku en Moquegue: Interacciones Regionales
y Colapso. Boletn de Arqueologa Pontica Uni-
versidad Catlica del Per 5: 169188. Lima.
Park, R. E.
1950 Human migration and the marginal man. In Race and
culture, edited by R. E. Park, 345356. Glencoe, IL:
The Free Press.
Parsons, J. R.
1968 An estimate of size and population for Middle
Horizon Tiahuanaco, Bolivia. American Antiquity
33:243245.
Perizonius, W. R. K.
1979 Non-metric traits: Sex difference and age depen-
dence. Journal of Human Evolution 8:679684.
Plog, S.
1983 Analysis of style in artifacts. Annual Review of Anthro-
pology 12:125142.
Ponce Sangins, C.
1972 Tiwanaku: Espacio, tiempo, y cultura, ensayo de sntesis
arqueolgica. Academia Nacional de Ciencias de
Bolivia, Publicacin 30. La Paz, Bolivia.
Posnansky, A.
1957 Tihuanacu: The cradle of American man, vols. 3 and 4.
La Paz: Ministerio de Educacin.
Prowse, T. L., and N. C. Lovell
1996 Concordance of cranial and dental morphological
traits and evidence for endogamy in ancient Egypt.
American Journal of Physical Anthropology 101:
237246.
Pucciarelli, H. M.
1975 The inuence of experimental deformation on neu-
rocranial wormian bones in rats. American Journal of
Physical Anthropology 41 (1): 2938.
Rasnake, R. N.
1988 Domination and cultural resistance: Authority and power
among Andean people. Durham, NC: Duke University
Press.
Rivera Casanovas, C. S.
1994 Chiji Jawira: Evidencias sobre la produccin de
cermica en Tiwanaku. Licenciatura thesis, Depart-
ment of Anthropology-Archaeology, Universidad
Mayor de San Andrs, La Paz, Bolivia.
Rosenthal, B.
1995 Iroquois false face masks. In Style, society, and person,
edited by C. Carr and J. E. Neitzel, 345367. New
York: Plenum Press.
Rothhammer, F., S. Quevedo, J. A. Cocilovo, and E. Llop
1984 Microevolution in prehistoric Andean popula-
tions: Chronologic non-metrical cranial variation
in northern Chile. American Journal of Physical
Anthropology 65 (2): 157162.
Royce, A. P.
1982 Ethnic identity: Strategies of diversity. Bloomington,
IN: Indiana University Press.
Rydn, S.
1957 Andean excavations 1: The Tiahuanaco era east of Lake
Titicaca. Ethnographical Museum of Sweden Mono-
graph Series 4. Stockholm.
1959 Andean excavations 2: Tupuraya and Cayhuasi, two Tia-
huanaco sites. Ethnographical Museum of Sweden
Monograph Series 6. Stockholm.
Saunders, S. R.
1978 The development and distribution of discontinuous mor-
phological variation of the human infracranial skeleton.
174 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
National Museums of Canada, Archaeological Sur-
vey of Canada, Paper 81. Ottawa.
1989 Nonmetric skeletal variation. In Reconstruction of life
from the skeleton, edited by M. Y. Iscan and K. A. R.
Kennedy, 95108. New York: Alan R. Liss.
Saunders, S. R., F. Popovich, and G. Thompson
1978 Family studies of non-metric skeletal traits: Atlas
bridging and clinoid bridging (meeting abstract).
American Journal of Physical Anthropology 44:203204.
Seddon, M. T.
1994 Excavations in the raised fields of the Ro Catari
Sub-Basin, Bolivia. Masters thesis, Department of
Anthropology, University of Chicago.
Shennan, S. J., ed.
1989 Archeological approaches to cultural identity. London:
Unwin Hyman.
Sofaer, J. A., P. Smith, and E. Kaye
1986 Afnities between contemporary and skeletal Jewish
and non-Jewish groups based on tooth morphol-
ogy. American Journal of Physical Anthropology 70:
265276.
Soto-Heim, P.
1987 Evolucin de deformaciones intencionales, tocados y
prcticas funerarias en la prehistoria de Arica, Chile.
Chungar 19:129213.
Squier, E. G.
1973 [1877] Peru: Incidents of travel and exploration in the
land of the Incas. New York: AMS Press for Peabody
Museum of Archaeology and Ethnology, Harvard
University, Cambridge, Massachusetts.
Stanish, C.
1992 Ancient Andean political economy. Austin: University
of Texas Press.
1994 The hydraulic hypothesis revisited: Lake Titicaca
basin raised elds in theoretical perspective. Latin
American Antiquity 5:312332.
Stewart, T. D.
1950 Deformity, trephining, and mutilation in South
American Indian skeletal remains. In Handbook of
South American Indians, vol. 6, Physical anthropology,
linguistics, and cultural geography of South American
Indians, edited by J. H. Steward, 4348. Bureau of
American Ethnology Bulletin 143. Washington,
D.C.: Smithsonian Institution.
Sutter, R. C.
1997 Dental variation and biocultural afnities among
prehistoric populations from the coastal valleys of
Moquegua, Per, and Azapa, Chile. Ph.D. disserta-
tion, Department of Anthropology, University of
Missouri, Columbia.
Trigger, B.
1986 Beyond history: The methods of prehistory. New York:
Holt, Rinehart and Winston.
Turner, C. G., II
1985 The dental search for Native American origins. In
Out of Asia, edited by R. Kirk and E. Szathmary,
3178. Canberra: Journal of Pacic History.
1987 Late Pleistocene and Holocene population history
of East Asia based on dental variation. American
Journal of Physical Anthropology 73 (3): 305321.
Turner, C. G., II, C. R. Nichol, and G. R. Scott
1991 Scoring procedures for key morphological traits of
the permanent dentition: The Arizona State Uni-
versity dental anthropology system. In Advances in
dental anthropology, edited by M. A. Kelley and C. S.
Larsen, 1332. New York: Wiley-Liss.
Turner, C. G., II, and G. R. Scott
1977 Dentition of Easter Islanders. In Orofacial growth
and development, edited by A. A. Dahlberg and T.
M. Graber, 229249. The Hague: Mouton.
Vaca de Castro, C.
1920 [1542] Discurso sobre la descendencia y gobierno de
los Incas. Declaracin de los quipocamayos a Vaca de
Castro. In Informaciones sobre el Antiguo Per (Crnicas
de 1535 a 1575), Coleccin de libros y documentos refer-
entes a la historia del Per, 2a serie, t. 3, edited by H.
H. Urteaga, transcribed by M. Jimnez de la Espada,
353. Lima: Imprenta y Librera San Mart.
Vranich, A.
1999 Interpreting the meaning of ritual spaces: The
temple complex of Puma Punku, Tiwanaku, Bolivia.
Ph.D. dissertation, Department of Anthropology,
University of Pennsylvania.
Weber, M.
1958 The city. Glencoe: The Free Press.
1979 Economy and society: An outline of interpretive sociology.
Berkeley: University of California Press.
Weiss, H. P.
1962 Tipologa de las deformaciones ceflicas de los anti-
guos Peruanos, segn la osteologa cultural. Revista
del Museo Nacional, Lima, Per 30:1542.
1972 Las deformaciones ceflicas intencionales como
factores de la arqueologa. In Actas y Memorias del
XXXIX Congreso Internacional de Americanistas, Lima
1970: Actas, Documentos, y Memorias, vol. I, edited by
R. Avalos de Matos and R. Ravines, 165180. Lima:
Instituto de Estudios Peruanos.
Wiessner, P.
1983 Style and social information in Kalahari San projec-
tile points. American Antiquity 48 (2): 253276.
B L O M 175
Winder, S. S.
1981 Infracranial non-metric variation: An assessment
of its value for biological distance analysis. Ph.D.
dissertation, Department of Anthropology, Indiana
University, Bloomington.
Yeatts, D. J.
1994 Economy, culture, and ethnicity in the vertical
Andes: The archaeological reconstruction of
Tiwanaku-Moquegua Valley interaction in the
Middle Horizon period in the Andes. Senior
Honors thesis, Department of Anthropology, Stan-
ford University.
176 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
APPENDIX 1: TRAITS AND CODING SCHEMA (BIODISTANCE)
NON-METRIC TRAITS USED IN THE STUDY
CRANIAL TRAITS: POSTCRANIAL TRAITS:
Accessory lesser palatine foramen Allens fossa
Apical bone Atlas lateral bridging*
Asterionic bone Atlas posterior bridge
Auditory exostoses Atlas retroarticular bridge
Bregmatic bone Atlas transverse foramen bridge*
Condylar canal Circumex sulcus*
Coronal ossicle Poiriers facet
Divided hypoglossal canal-internal* Septal aperture*
Divided hypoglossal canal-within canal Squatting facet-lateral*
Epipteric bone Squatting facet-medial*
Foramen ovale incomplete Sternal foramen*
Foramen spinosum incomplete Suprascapular foramen/notch
Inca bone Supratrochlear spur
Infraorbital suture Talar articulation shape*
Lambdoid ossicle Third trochanter*
Mandibular torus Trochlear notch form*
Mastoid foramen Vastus notch*
Mental foramen*
Metopic suture* DENTAL TRAITS:
Multiple infraorbital foramina Groove pattern-lower molars [M1*]
Mylohyoid bridge-center of groove* Cusp number-lower molars [M1*]
Mylohyoid bridge-mandibular foramen* Root number-lower molars [M1*]
Ossicle in occipito-mastoid suture Root number-upper 1st premolar*
Palatine torus Protostylid-lower molars [M1*]
Parietal foramen Cusp 5-lower molars [M1*]
Parietal notch bone Cusp 6-lower molars [M1*]
Pterygo-alar bridge Cusp 7-lower molars [M1*]
Pterygo-spinous bridge Tomes root-lower 1st premolar*
Rocker mandible Shoveling-upper incisors [I1*]
Sagittal ossicle Double shoveling-upper incisors [I1*]
Superior sagittal sulcus turns left* Peg-shaped incisors-upper lateral incisor*
Suprameatal pit* Winging-upper central incisors*
Supraorbital foramen* Hypocone-upper molars [M1*]
Supratrochlear foramen* Cusp 5 (metaconule)-upper molars [M1*]
Tympanic dihiscence* Carabellis trait-upper molars [M1*]
Zygomatico-facial foramen Enamel extensions-upper molars [M1*]
*Traits used in nal analysis.
B L O M 177
CODING SCHEMA CRANIAL
NON-METRIC TRAITS
METOPIC SUTURE:
1. absent
2. partial
3. complete
9. unobservable
SUPRAORBITAL NOTCH:
1. absent
2. <1/2 occluded by spicules
3. >1/2 occluded by spicules
4. present, degree unknown
5. multiple notches
9. unobservable
SUPRAORBITAL FORAMEN:
1. absent
2. present
3. multiple foramina
9. unobservable
SUPRATROCHLEAR NOTCH:
1. absent
2. without spur
3. with spurs
9. unobservable
SUPRATROCHLEAR FORAMEN:
1. absent
2. present
9. unobservable
INFRAORBITAL SUTURE:
1. absent
2. partial
3. complete
9. unobservable
MULTIPLE INFRAORBITAL FORAMINA:
1. absent
2. internal division only
3. two distinct foramina
4. > two distinct foramina
9. unobservable
ZYGOMATICO-FACIAL FORAMEN:
1. absent
2. 1 large
3. 1 large plus smaller f.
4. 2 large
5. 2 large plus smaller f.
6. 1 small
7. multiple small
8. multiple large
9. unobservable
PARIETAL FORAMEN:
1. absent
2. present, on parietal
3. present, on sutural
9. unobservable
SUTURAL BONES:
1. absent
2. present
9. unobservable
a. epiteric bone
b. coronal ossicle
c. bregmatic bone
d. sagittal ossicle
e. apical bone (lambda)
f. lambdoid ossicle
g. asterionic bone
h. ossicle in occipito-mastoid suture
i. parietal notch bone
INCA BONE:
1. absent
2. complete, single bone
3. bipartite
4. tripartite
5. partial
9. unobservable
CONDYLAR CANAL:
1. absent
2. present
9. unobservable
DIVIDED HYPOGLOSSAL CANAL:
1. absent
2. partial, internal surface
3. partial, within canal
4. complete, internal surface
5. complete, within canal
9. unobservable
FLEXURE OF SUPERIOR SAGITTAL
SULCUS:
1. right
2. left
3. bifurcate
4. right with elevation
5. left with elevation
9. unobservable
FORAMEN OVALE INCOMPLETE:
1. absent
2. partial formation
3. no denition of foramen
9. unobservable
FORAMEN SPINOSUM INCOMPLETE:
1. absent
2. partial formation
3. no denition of foramen
9. unobservable
PTERYGO-SPINOUS BRIDGE:
1. absent
2. trace (spicule only)
3. partial bridge
4. complete bridge
5. 2 spicules bilateral to foramen
9. unobservable
PTERYGO-ALAR BRIDGE:
1. absent
2. trace (spicule only)
3. partial bridge
4. complete bridge
9. unobservable
TYMPANIC DIHISCENCE:
1. absent
2. foramen only
3. full defect present
9. unobservable
AUDITORY EXOSTOSES:
1. absent
2. <1/3 canal occluded
3. 1/3. 2/3 canal occluded
4. >2/3 canal occluded
9. unobservable
SUPRAMEATAL PIT:
1. absent
2. small pit
3. large, shelf like
9. unobservable
ACCESSORY LESSER PALATINE
FORAMEN:
1. absent
2. present
9. unobservable
b. Number:
1. absent
2. 1
3. 2
4. > 2
9. unobservable
MANDIBULAR TORUS:
1. absent
2. trace (can palpate)
3. moderate; elevation between 2-5 mm
4. marked; elevation > 5 mm
9. unobservable
ROCKER MANDIBLE:
1. absent
2. present
9. unobservable
MENTAL FORAMEN:
1. absent
2. 1
3. 2
4. more than 2
9. unobservable
MYLOHYOID BRIDGE:
a. Location:
1. absent
2. near mandibular foramen
3. center of groove
4. both bridges described in 1) and 2)
with hiatus
5. both bridges described in 1) and 2),
no hiatus
9. unobservable
b. Degree:
1. absent
2. partial (denite spicule)
3. compete
9. unobservable
178 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
CODING SCHEMA POSTCRANIAL NON-METRIC TRAITS
FEMUR POIRIERS FACET:
1. Absent
2. Slight
3. Marked
9. Unobservable
FEMUR ALLENS FOSSA:
1. Absent
2. Small
3. Large
9. Unobservable
TIBIA SQUATTING FACETS:
1. Absent
2. Medial
3. Lateral
4. Medial and lateral
9. Unobservable
CALCANEUS SHAPE OF TALAR ARTICULAR
SURFACES:
1. Anterior absent
2. Ovoid
3. Hourglass
4. 2 facets
5. Hourglass with slight division
9. Unobservable
SCAPULA SUPRASCAPULAR FORAMEN/NOTCH FORM:
1. Absent
2. Notch with spurs
3. Foramen
9. Unobservable
SCAPULA CIRCUMFLEX SULCUS:
1. Absent
2. Present
9. Unobservable
FEMUR 3RD TROCHANTER:
1. Absent
2. Slight
3. Marked
9. Unobservable
PATELLA VASTUS NOTCH:
1. Absent
2. Present
9. Unobservable
HUMERUS SUPRATROCHLEAR SPUR:
1. Absent
2. Small (<1cm)
3. Large (>1cm)
9. Unobservable
STERNUM STERNAL FORAMEN:
1. Absent
2. Present
9. Unobservable
ULNA TROCHLEAR NOTCH FORM:
1. Absent
2. 2 facets
3. one facet with large division
9. Unobservable
3RD 7TH CERVICAL VERTEBRAE TRANSVERSE
FORAMEN BIPARTITE
# Present/# Observable
1
ST
CERVICAL VERTEBRA POSTERIOR BRIDGE:
1. Absent
2. Spurs
3. Bridge
9. Unobservable
1
ST
CERVICAL VERTEBRA RETROARTICULAR BRIDGE:
1. Absent
2. Spurs
3. Bridge
9. Unobservable
B L O M 179
CODING SCHEMA DENTAL NON-METRIC TRAITS
LOWER MOLARS:
PROTOSTYLID
1. No expression of any sort. Buccal surface is smooth.
2. A pit occurs in the buccal groove.
3. Buccal groove is curved distally.
4. A faint secondary groove extends mesially from the
buccal groove.
5. Secondary groove is slightly more pronounced.
6. Secondary groove is stronger and can easily be seen.
7. Secondary groove extends across most of the buccal
surface of cusp 1. This is considered a weak or small
cusp.
8. A cusp with a free apex occurs.
9. Unobservable
CUSP 5
1. Absence. Molar has only 4 cusps.
2. Cusp 5 is present and very small.
3. Cusp 5 is small
4. Cusp 5 is medium sized.
5. Cusp 5 is large
6. Cusp 5 is very large.
7. Cusp 6 is present.
9. Unobservable
CUSP 6
1. Cusp 6 is absent.
2. Cusp 6 is much smaller than cusp 5.
3. Cusp 6 is smaller than cusp 5.
4. Cusp 6 is equal in size to cusp 5.
5. Cusp 6 is larger than cusp 5.
6. Cusp 6 is much larger than cusp 5.
9. Unobservable
CUSP 7
1. No occurrence of cusp 7.
2. Faint cusp is present. Two weak lingual grooves are
present instead of one.
3. A faint tipless cusp 7 occurs displaced as a bulge on the
lingual surface of cusp 2.
4. Cusp 7 is small.
5. Cusp 7 is medium sized.
6. Cusp 7 is large.
9. Unobservable
ROOT NUMBER
1. One Root: root tip may be bifurcated. If tips are free
for more than one-fourth to one-third of the total root
length, score as two roots.
2. Two Roots: Two separate roots exist for at least 1/4 to
1/3 the total root length.
3. Three Roots: A third (supernumerary) root is present
on the distolingual aspect.
9. Unobservable
LOWER FIRST PREMOLARS:
ROOT NUMBER (TOMES ROOT)
1. Developmental grooving is absent or, if present,
shallow with rounded rather than V-shaped
indentation.
2. Developmental groove is present and has a shallow
V-shaped cross-section.
3. Developmental groove is present and had a moderately
deep V-shaped cross-section.
4. Developmental groove is present, V shaped, and deep.
5. Developmental grooving is deeply invaginated on both
the mesial and distal borders.
6. Two free roots are present. They are separate for at
least one-quarter to one-third of the total root length.
9. Unobservable
UPPER CENTRAL INCISORS:
SHOVELING
1. None: Lingual surface is essentially at.
2. Faint: Vary slight elevations of mesial and distal
aspects of lingual surface can be seen and palpated.
3. Trace: Elevations are easily seen.
4. Semi-shovel: Stronger ridging is present and there is a
tendency for ridge convergence at the cingulum.
5. Semi-shovel: Convergence and ridging are stronger
than in grade 3.
6. Shovel: Strong development of ridges, which almost
contact at the cingulum.
7. Marked Shovel: Strongest development. Mesial and
distal lingual ridges are sometimes in contact at the
cingulum.
8. Barrel (U12 only): Expression exceeds grade 6.
9. Unobservable
DOUBLE SHOVELING
1. None: Labial surface is smooth.
2. Faint: Mesial and distal ridging can be seen in strong
contrasting light. Distal ridge may be absent in this
and stronger grades.
3. Trace: Ridging is more easily seen and palpated.
4. Semi-Double Shovel: Ridging can be readily palpated.
5. Double Shovel: Ridging is pronounced on at least one-
half of the total crown height.
6. Pronounced Double Shovel: Ridging is very
prominent and may occur from the occlusal surface to
the CEJ.
7. Extreme Double Shovel: Very pronounced ridging.
9. Unobservable
WINGING
1. Absent.
2. Bilateral Winging: Central incisors are rotated
mesiolingually. When the angle formed is greater
than 20 degrees, it is classed as 1A; when less than 20
degrees, 1B.
3. Unilateral Winging: Only one of the incisors is
rotated.
4. Straight: Both teeth form a straight labial surface or
follow the curvature of the arcade.
5. Counter Winging: One or both teeth are rotated
distolingually.
9. Unobservable
180 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
UPPER LATERAL INCISORS:
PEG-SHAPED INCISORS
1. Normal-sized incisor.
2. Incisor reduced in size, but having normal crown form.
3. Peg-shaped incisor as dened above.
9. Unobservable
UPPER MOLARS:
CARABELLIS TRAIT
1. Mesiolingual aspect of cusp 1 is smooth.
2. Groove present.
3. Pit present.
4. A small Y-shaped depression is present.
5. A large y-shaped depression is present.
6. A small cusp without a free apex occurs. The distal
border of the cusp does not contact the lingual groove
separating cusps 1 and 4.
7. A medium-sized cusp with an attached apex making
contact with the medial lingual groove is present.
8. A large free cusp is present
9. Unobservable
CUSP 5 (METACONULE)
1. Site of cusp 5 is smooth, with only a single distal
groove separating cusps 3 and 4.
2. Faint cuspule present.
3. Trace cuspule present.
4. Small cuspule present
5. Small cusp present.
6. Medium-sized cusp present.
9. Unobservable
ENAMEL EXTENSIONS
1. Enamel border is straight, or, rarely, curved toward the
crown. Score any extension not attached to the crown
as absent.
2. A faint, approximately 1.0mm long extension
projecting toward and along the root.
3. A medium-sized, approximately 2.0 mm long
extension.
4. A lengthy extension, generally >4.00 mm in length. It
may extend all the way to the root bifurcation.
9. Unobservable
UPPER FIRST PREMOLARS:
ROOT NUMBER
1. One root. Tip may be bid for less than 1/3 of its
length.
2. Two roots. Separate roots must be greater than one-
quarter to one-third of the total root length.
3. Three roots. Length dened as in grade 2.
9. Unobservable
B L O M 181
RECODING INTO PRESENT/ABSENT FOR NON-METRIC TRAITS
Skeletal Trait Recoding
Trait Name Present Absent
Accessory Lesser Palatine Foramen 2 1
Allens Fossa 3 2, 1
Apical Bone 2 1
Asterionic Bone 2 1
Atlas Lateral Bridging 2, 3 1
Atlas Posterior Bridge 2, 3 1
Atlas Retroarticular Bridge 2, 3 1
Atlas Transverse Foramen Bridge 1+/1+ 0/1+
Auditory Exostoses 2 1
Bregmatic Bone 2 1
Circumex Sulcus 2 1
Condylar Canal 2 1
Coronal Ossicle 2 1
Divided Hypoglossal Canal-Internal 2, 4 1
Divided Hypoglossal Canal-Within Canal 3, 5 1
Epiteric bone 2 1
Flexure of Superior Sagittal Sulcus-Left 2, 3, 5 1, 4
Foramen Ovale Incomplete 2 1
Foramen Spinosum Incomplete 2 1
Inca Bone 2-5 1
Infraorbital Suture 2 1
Lambdoid Ossicle 2 1
Mandibular Torus 3, 4 1, 2
Mastoid Foramen 1
Mental Foramen 3, 4 1, 2
Metopic Suture 3 1, 2
Multiple Infraorbital Foramina 2 1
Mylohyoid Bridge Center of Groove location=2+
number=1+
1
Mylohyoid Bridge Near Mandibular Foramen location=2, 4, 5
number=2+
1
Ossicle in Occipito-Mastoid Suture 2 1
Palatine Torus 2 1
Parietal Foramen 2 1
Parietal Notch Bone 2 1
Poiriers Facet 2, 3 1
Pterygo-alar Bridge 3, 4 1, 2
Pterygo-spinous Bridge 3, 4 1, 2
Rocker Mandible 2 1
Sagittal Ossicle 2 1
Septal Aperture 2, 3 1
Squatting Facet-Lateral 2 1
Squatting Facet-Medial 2 1
Sternal Foramen 2 1
Supratrochlear Foramen 2 1
Supratrochlear Spur
Suprameatal Pit 2 1
Supraorbital Foramen 2 1
Suprascapular Foramen/Notch Form 2, 3 1
Supratrochlear Spur-Humerus 2, 3 1
Talar Articulation 2 facets 4 2, 3, 5
Third Trochanter 2, 3 1
Trochlear Notch Form 2 1, 3*
Tympanic Dihiscence 3 1, 2
Vastus Notch 2 1
Zygomatico-facial Foramen 3-5, 7 1, 2, 6
182 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Dental Trait Recoding
Arcade Trait Present Absent Null
Lower Groove Pattern Y X,+,W
Cusp#-M2&3 4 5, 6
Cusp#-M1 6 4, 5
Protostylid 3-8 1, 2
Cusp5 (Hypoconulid) 2-6 1 7
Cusp6 (Entoconulid) 2-6 1
Cusp7 (Metaconulid) 2-6 1
Root#-M2&3 2 1, 3
Root#-M1 3 1, 2
Root#-PM1
Upper Root#-PM1 1 2, 3
Hypocone 6, 7 1-5
Metaconule 2-6 1
Carabellis Cusp 3-8 1, 2
Enamel Extension 3, 4 1, 2
Shoveling 3-8 1, 2
Double shoveling 3-7 1, 2
Winging 2, 3 1, 4, 5
Peg-shaped 3 1, 2
183
INTRODUCTION
The concept of ethnicity is widely recognized as one
that is both difcult to dene and operationalize, even
among living people. While much disagreement exists
on exactly what denes ethnicity, most scholars agree
that ethnic identity often involves assigning importance
to perceived afnities among individuals as well as some
sense of differences among groups (Barth 1969; Carter
Bentley 1987; Drummond 1980; Southall 1976). Fur-
ther, group afnities and differences are often expressed
culturally through both material objects and shared,
group-specic behavior (Carter Bentley 1987).
APPROACHES TO PREHISTORIC ETHNICITY IN
THE SOUTH-CENTRAL ANDES
As is evident by the contributions of this volume, a num-
ber of archaeologists and bioarchaeologists working in
the south-central Andes have explored methodologies
for identifying prehistoric ethnic groups. More speci-
cally, archaeologists working within the south-central
Andes have explored material correlates of ethnicity,
largely due to the stated goal of identifying ethnic alti-
plano colonies within the coastal valleys of the region.
The identication of ethnic groups in the archaeologi-
cal record has traditionally been based upon variation
among artifacts and features that are thought to have
been contemporaneous.
Stanish (1989b) notes that archaeological investi-
gations in the south-central Andes have traditionally
relied upon what he refers to as an artifact-based
approach to determine both the cultural and biologi-
cal afliation among sites from different regions. He
denes this approach as one that relies upon the pres-
ence of particular classes of artifacts (generally ceramics
and other highly valued objects from funerary and cer-
emonial contexts) but fails to control for the context
where they are found. These artifacts are then used to
determine a given sites cultural afliation. However,
as Stanish (1989a, 1989b) and others (Aldenderfer and
Stanish 1993) correctly indicate, this approach does
not distinguish among the material correlates of trade,
CHAP T E R 1 1
A Bioarchaeological Assessment
of Prehistoric Ethnicity
among Early Late Intermediate
Period Populations of the
Azapa Valley, Chile
RI CHARD C. S UTTER
184 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
ideology, and colonization; one of the greatest problems
with an artifact-based methodology is that artifacts such
as ceramics, textiles, and metal objects are highly trans-
portable and can appear in archaeological contexts far
outside their range of production.
Because altiplano colonists are expected to defend
and display their ethnic afliation with the homeland,
ethnicity has been a concept of primary importance
for identifying altiplano colonies in the archaeological
record. A variety of artifacts, features, and cultural prac-
tices believed to be reliable indicators of ethnicity have
been used to test the presence of ethnically altiplano
colonists in the coastal valleys of the Andes. These
include ceramics (Owen 1993), textiles (Cassman 1997;
Oakland 1992), domestic architecture (Aldenderfer and
Stanish 1993; Stanish 1989b), mortuary practices (Bui-
kstra 1995), and cranial deformations (Lozada, Blom,
and Buikstra 1996), all of which have been used to test
the presence of altiplano colonies. The interest, in these
cases, has been to identify material objects, features, or
other indicators of culturally mediated behavior that
reect stylistic similarities among assemblages from the
coastal valleys with those of known altiplano origin.
While not always explicitly stated, the goal of archae-
ologically based methodologies of ethnicity is not only
to identify cultural afliation with the altiplano, but also
to infer genetic relatedness with altiplano populations.
However, if the goal is to identify genetic relatedness
among coastal and highland populations sharing cultural
expressions of group identity, this may be confounded
by a variety of factors. Although recent investigations
that explore prehistoric ethnicity are more explicit in
their methodology, they largely ignore genetic rela-
tions among the groups being studied (although see
Haydon 1993; Lozada, Haydon, and Buikstra 1993;
Sutter 1997, 2000).
An implicit assumption of many archaeological
investigations that examine prehistoric ethnicity is
that spatially associated contemporaneous stylistic
differences among cultural practices (i.e., features and
artifact assemblages, bodily modication) represent
prehistoric multiethnic settlement. However, the
usefulness of archaeologically based methodologies is
unclear because there is no a priori reason to assume
that any given culturally based behavior (i.e., artifact
manufacture and use, archaeological features, cranial
deformation, etc.) will be an indicator of group afli-
ation. This is because group formation and afliation
have been shown to often be situational (Geary 1983).
Indeed, there may be different social contexts that evoke
different group afliations (Berreman 1972; Carter
Bentley 1987; Cohen 1981), and instances when indi-
viduals will hold multiple ethnic identities that are not
necessarily mutually exclusive (Keyes 1976). For these
reasons, ethnicity has remained a difcult concept to
dene (Barth 1969; Carr and Neitzel 1995) and presents
unique challenges for those trying to apply the concept
to the archaeological record (Aldenderfer and Stanish
1993; Bawden 1993; Bermann 1994).
AN ALTERNATIVE APPROACH
BIOARCHAEOLOGY AS METHODOLOGY
Because no class of objects will always be directly cor-
related with ethnic afliation, or, necessarily, with the
genetic composition of a given population, the concept
of ethnicity as applied to the archaeological record can-
not, by itself, satisfactorily address the larger problem
of identifying the source(s) for the formation of ethnic
groups. For these reasons, a bioarchaeological method-
ology has some advantages over strictly archaeologically
based methodologies, if the relationship between genetic
relatedness and culturally expressed group identity (i.e.,
ethnicity) can be identied. This is an important meth-
odological distinction because groups of people can be
genetically related (i.e., have recent common ancestors),
yet actively express group distinctions through mate-
rial culture. On the other hand, genetically unrelated
populations with distinct cultural histories can share
economic orientation, ideology, religion, and ethnic or
political afliation (Sutter 1997, 2000).
Within this paper I utilize the concept of biocultural
groups. Biocultural groups are operationally dened here
as groups that demonstrate ethnic afliation as indicated
by culturally based behavior, genetic relatedness, and
shared economic activities and interests (Carter Bentley
1987). The assumption of this approach is that each of
these factors will be inuenced, to varying degrees, by
culturally dened group afliation (gure 1).
For this study, conscious stylistic (emblemic) expres-
sion of group identity is inferred by examining grave
offerings and cranial deformations, while shared socio-
economic activities, interests, and diet are inferred using
both dental pathologies and grave furniture. Genetic
relatedness is explored using both dental and cranial
epigenetic traits. This methodology is illustrated using
four early Late Intermediate period mortuary popula-
tions from the Azapa Valley, Chile.
The goals of this paper are to uncover which of
those variables are related to ethnic variation and to
S U T T E R 185
identify the basis of ethnic variation among these Late
Intermediate period mortuary populations. In doing
so, I hope to illustrate the usefulness of the proposed
bioarchaeological methodology.
ESTABLISHING LATE INTERMEDIATE PERIOD
ETHNICITY WITHIN THE AZAPA VALLEY
A basic, yet far from trivial task for all studies attempt-
ing to identify ethnicity-based variation in the
archaeological record is to establish the contempora-
neity among apparently distinct cultural groups. The
culture history for the Azapa Valley, especially for the
terminal Middle Horizon and early Late Intermedi-
ate period (AD 9501476), as reported to date, is at
best confusing. This is especially true for the Cabuza,
Maitas-Chiribaya, and San Miguel ceramic traditions.
Most scholars working in the Azapa Valley argue that
the presence of Middle Horizon Tiwanaku V, Loreto
Viejo, Sobraya, Cabuza, and Maitas-Chiribaya ceramics
from mortuary contexts within the same cemetery is
evidence of multiethnic settlement of the Azapa Valley
by altiplano colonists. However, many archaeologists
working in the Moquegua Valley, Peruwhere cultural
and ceramic traditions are similar to those for the Azapa
Valleyhave commented on the wide discrepancies
among the dates for the cultural developments within
these two valleys (Goldstein 1995; Owen 1993; Sut-
ter 1997). For example, the Chilean Sobraya, Cabuza,
and Maitas-Chiribaya ceramic traditions demonstrate
clear formal and stylistic similarities to Moquegua Val-
ley Tumilaca, Ilo-Tumilaca, Ilo-Cabuza, and Chiribaya
ceramic traditions recovered from stratied post-Tiwa-
naku Late Intermediate period contexts.
The Late Intermediate period is defined by the
decline of Tiwanakus inuence in the region. Tiwan-
akus collapse appears to have begun in the periphery
rst and progressed toward the Tiwanaku heartland.
While Tiwanakus inuence may have persisted until
AD 1200 in the altiplano (Kolata 1993), evidence from
the Moquegua and Azapa valleys suggests Tiwanakus
inuence ended in those areas sometime prior to AD
950 (Goldstein 1993, 1995). Following its demise in the
coastal valleys of the south-central Andes, the cultural
inuence of Tiwanaku was quickly replaced by distinc-
tive regional architecture, ceramics, burial practices,
and textiles.
While scholars working in the Azapa Valley believe
that Cabuza and Maitas-Chiribaya ceramics are Middle
Horizon ceramic traditions, a critical evaluation of both
stylistic and chronometric evidence clearly indicates
these traditions largely postdate the Middle Horizon.
The archaeological and radiometric evidence I present
here indicates that these ceramic traditions have been
chronologically misplaced. Many of the radiocarbon
dates for the Azapa Valley are presented here with
their calibrations for the rst time. This comparison
provides essential information that permits a critical
evaluation of culture history for this valley. To ensure
comparability of calibrated dates presented here, I
used MacCALIB 3.0.3A (Stuiver and Reimer 1993b)
to calibrate all radiometric dates using Method A. The
Figure 1: Components of ethnicity and methods used by this study to infer each component.
186 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
calibrated dates produced using Method A represent
the intercept of the radiocarbon age with the calibration
curve (Stuiver and Reimer 1993a).
With the exception of a single calibrated radiocarbon
date of AD 540 and a second date of AD 310 obtained
through thermoluminesence (Schiappacasse et al. 1991:
45), the remaining thermoluminesence and sixteen cali-
brated radiocarbon dates for Cabuza range between AD
1020 and 1430 (table 1). While the two earlier dates
clearly allow for some overlap with Tiwanaku, the
majority of these dates rmly place the Cabuza tradition
contemporaneous with early Late Intermediate period
Tumilaca and Ilo-Tumilaca/Cabuza ceramic traditions
of the Moquegua Valley (Owen 1993). Indeed, although
most of the dates from Cabuza range between AD 1020
and AD 1250, some of the Cabuza dates from textiles
are considerably more recent than is expected for this
cultural tradition.
The Middle Horizon antiquity of the Chilean Maitas-
Chiribaya ceramic tradition is also doubtful for the same
reasons. Muoz (1983:83) reports ten radiocarbon dates
from Maitas-Chiribaya funerary contexts that, when
calibrated, range between AD 890 and 1030. However,
nineteen additional radiocarbon dates and more than
fteen thermoluminesence dates for Maitas-Chiribaya
range between AD 950 and 1440 (Schiappacasse et al.
1991:45), clearly placing the Maitas-Chiribaya tradition
in the Late Intermediate period (table 2).
The black paint on white slip San Miguel ceramic
tradition is closely related to the middle and terminal
Late Intermediate period Pocoma and Gentilar ceramic
traditions (Espoueys et al. 1994). The calibrations for
the six reported radiocarbon dates for San Miguel
range between AD 1030 and 1380 (table 3), making
this tradition contemporaneous with the Cabuza and
Maitas-Chiribaya cultural traditions. The forms and
pastes of Azapa Valley San Miguel ceramics are nearly
identical to Maitas-Chiribaya ceramics (Mariela San-
tos, personal communication 1994), also suggesting
that these ceramics are contemporaneous. Further,
the co-occurrence in tombs of San Miguel ceramics
with Cabuza and Maitas-Chiribaya ceramics is also an
independent, albeit ambiguous, indication that these
ceramic traditions are contemporaneous (Espinoza et
al. 1994). This is not a widely held opinion of archae-
ologists working in northern Chile, who see the San
Miguel tradition as being derived from the Maitas-
Chiribaya tradition (Espoueys et al. 1994; Muoz and
Focacci 1985). However, when critically evaluated, the
archaeological record for the Azapa Valley indicates the
coexistence of the Cabuza, Maitas-Chiribaya, and San
Miguel ceramic traditions during the Late Intermediate
period. It is important to emphasize that none of these
three ceramic traditions considered here has a known
altiplano counterpart.
MATERIALS AND METHODS
I. Grave Good Analysis
Mortuary practices are well understood for Late
Intermediate period cultural traditions within the Azapa
Valley. This investigation examines human burials and
their associated goods from Late Intermediate period
sites Playa Miller-4, Azapa-140, Azapa-71, and Azapa-
6 of the Azapa Valley, Chile (gure 2). Grave good
data discussed by this investigation are based upon the
examination of textiles by Cassman (1997), the con-
sultation of site reports and eld notes (Focacci 1961,
1968, 1969a, 1969b, 1982, 1993; Muoz and Focacci
1985), and a recently compiled grave good database
for the mortuary populations examined by this study
(Espinoza et al. 1994).
1) Playa Miller-4. Playa Miller-4 is a Late Intermedi-
ate period cemetery site associated with a shing
village. The cemetery is located approximately
50 m down the coast from the Formative period
Playa Miller-7 site and is adjacent to the contem-
poraneous cemetery sites Playa Miller-2, 6, and 9
(Focacci 1969b, 1974, 1982).
These nearby sites represent different areas of the
same prehistoric cemetery that were excavated during
subsequent field seasons. Only those remains from
Playa Miller-4 were available for this study. With the
exception of a few brief remarks in publications by
Focacci (1980), Muoz (1983), and Soto-Heim (1987),
relatively little has been published concerning Playa
Miller-4. Tombs at Playa Miller-4 were pits excavated
into the sand. Graves were frequently lined with two or
three rock slabs to support the walls of the tomb. Playa
Miller-4 tombs contained single, exed burials placed
in their graves in a seated position. Grave lots indicate
that maritime economic activities were practiced by
individuals interred at Playa Miller-4. The remains and
grave goods of forty-one individuals were examined for
this site, and were primarily associated with ceramics of
the San Miguel tradition.
2) Azapa-140. The cemetery site Azapa-140 is located
12.5 km inland from the coast, and a few meters
east of the habitational site Azapa-11, also known
as Pucar San Lorenzo (Muoz and Focacci 1985).
S U T T E R 187
T
a
b
l
e
1
.
R
a
d
i
o
c
a
r
b
o
n
d
a
t
e
s
f
o
r
t
h
e
A
z
a
p
a
V
a
l
l
e
y
C
a
b
u
z
a
c
e
r
a
m
i
c
t
r
a
d
i
t
i
o
n
.
U
n
c
o
r
r
e
c
t
e
d
S
i
t
e
C
o
n
t
e
x
t
C
u
l
t
u
r
e
L
a
b
I
D
S
a
m
p
l
e
C
1
4
D
e
n
d
o
-
C
a
l
i
b
r
a
t
e
d
R
e
f
e
r
e
n
c
e
A
Z
-
7
1
T
-
3
3
2
C
a
b
u
z
a
B
e
t
a
-
7
8
9
6
3
T
e
x
t
i
l
e
4
9
0
6
0
A
D
1
4
0
7
(
1
4
3
3
)
1
4
4
9
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
2
1
C
a
b
u
z
a
B
e
t
a
-
7
8
9
5
3
T
e
x
t
i
l
e
6
5
0
6
0
A
D
1
2
9
0
(
1
3
0
5
,
1
3
6
7
,
1
3
7
3
)
1
3
9
8
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
2
2
8
C
a
b
u
z
a
B
e
t
a
-
7
8
9
6
0
T
e
x
t
i
l
e
6
6
0
7
0
A
D
1
2
8
6
(
1
3
0
2
)
1
3
9
8
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
1
1
2
A
C
a
b
u
z
a
B
e
t
a
-
7
8
9
5
5
T
e
x
t
i
l
e
6
7
0
5
0
A
D
1
2
8
8
(
1
3
0
0
)
1
3
9
1
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
2
1
4
C
a
b
u
z
a
B
e
t
a
-
7
8
9
5
2
T
e
x
t
i
l
e
8
0
0
7
0
A
D
1
1
9
2
(
1
2
5
3
)
1
2
8
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
2
7
8
A
C
a
b
u
z
a
B
e
t
a
-
8
0
6
3
4
T
e
x
t
i
l
e
8
1
0
7
0
A
D
1
1
7
4
(
1
2
4
5
)
1
2
8
4
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
1
8
7
C
a
b
u
z
a
B
e
t
a
-
7
8
9
5
9
T
e
x
t
i
l
e
8
2
0
4
0
A
D
1
2
1
3
(
1
2
2
9
)
1
2
7
7
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
1
4
7
C
a
b
u
z
a
B
e
t
a
-
7
7
2
2
2
T
e
x
t
i
l
e
8
3
0
5
0
A
D
1
1
7
4
(
1
2
2
5
)
1
2
7
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
1
7
3
C
C
a
b
u
z
a
B
e
t
a
-
7
7
2
2
3
T
e
x
t
i
l
e
8
5
0
4
0
A
D
1
1
6
8
(
1
2
1
8
)
1
2
4
9
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
9
9
B
C
a
b
u
z
a
B
e
t
a
-
7
8
9
5
4
T
e
x
t
i
l
e
8
5
0
5
0
A
D
1
1
6
5
(
1
2
1
8
)
1
2
5
5
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
1
7
4
C
a
b
u
z
a
B
e
t
a
-
7
8
9
5
8
T
e
x
t
i
l
e
8
5
0
5
0
A
D
1
1
6
5
(
1
2
1
8
)
1
2
5
5
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
2
4
2
C
a
b
u
z
a
B
e
t
a
-
7
8
9
6
1
T
e
x
t
i
l
e
8
5
0
6
0
A
D
1
1
6
2
(
1
2
1
8
)
1
2
6
7
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
1
T
-
5
2
C
a
b
u
z
a
-
M
u
s
c
l
e
8
7
0
8
0
A
D
1
0
4
1
(
1
1
9
5
)
1
2
6
6
M
u
s
e
o
S
a
n
M
i
g
u
e
l
d
e
A
z
a
p
a
L
a
b
R
e
c
o
r
d
s
A
Z
-
1
4
1
T
-
2
4
C
a
b
u
z
a
I
-
1
3
,
7
8
0
M
u
s
c
l
e
9
3
0
8
0
A
D
1
0
1
9
(
1
0
5
2
,
1
0
8
5
,
1
1
2
1
,
1
1
3
9
,
1
1
5
6
)
1
2
1
8
M
u
s
e
o
S
a
n
M
i
g
u
e
l
d
e
A
z
a
p
a
L
a
b
R
e
c
o
r
d
s
A
Z
-
1
4
1
T
-
4
7
C
a
b
u
z
a
-
M
u
s
c
l
e
9
8
0
8
0
A
D
9
9
6
(
1
0
2
8
)
1
1
6
5
M
u
s
e
o
S
a
n
M
i
g
u
e
l
d
e
A
z
a
p
a
L
a
b
R
e
c
o
r
d
s
A
Z
-
7
1
T
-
1
4
0
C
a
b
u
z
a
B
e
t
a
-
7
8
9
5
6
T
e
x
t
i
l
e
1
0
1
0
8
0
A
D
9
7
8
(
1
0
2
0
)
1
1
5
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
6
T
-
2
5
C
a
b
u
z
a
-
S
o
b
r
a
y
a
G
a
K
5
8
1
6
C
a
m
o
t
e
1
5
7
0
6
5
A
D
4
2
1
(
5
3
5
)
5
9
4
F
o
c
a
c
c
i
1
9
8
2
:
7
4
188 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
T
a
b
l
e
2
.
R
a
d
i
o
c
a
r
b
o
n
d
a
t
e
s
f
o
r
t
h
e
A
z
a
p
a
V
a
l
l
e
y
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
c
e
r
a
m
i
c
t
r
a
d
i
t
i
o
n
.
U
n
c
o
r
r
e
c
t
e
d
S
i
t
e
C
o
n
t
e
x
t
C
u
l
t
u
r
e
L
a
b
I
D
S
a
m
p
l
e
C
1
4
D
e
n
d
o
-
C
a
l
i
b
r
a
t
e
d
R
e
f
e
r
e
n
c
e
A
Z
-
1
4
0
T
-
6
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
6
2
7
T
e
x
t
i
l
e
4
5
0
5
0
A
D
1
4
3
0
(
1
4
4
4
)
1
4
7
3
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
4
4
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
9
6
7
T
e
x
t
i
l
e
5
0
0
6
0
A
D
1
4
0
5
(
1
4
3
1
)
1
4
4
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
1
0
0
M
a
i
t
a
s
-
C
h
r
i
b
a
y
a
B
e
t
a
-
7
8
6
3
1
T
e
x
t
i
l
e
5
3
0
5
0
A
D
1
4
0
0
(
1
4
1
5
)
1
4
3
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
2
8
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
6
2
8
T
e
x
t
i
l
e
5
6
0
5
0
A
D
1
3
2
1
(
1
4
0
5
)
1
4
2
7
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
7
1
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
6
3
0
T
e
x
t
i
l
e
6
7
0
6
0
A
D
1
2
8
6
(
1
3
0
0
)
1
3
9
3
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
1
0
1
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
9
6
8
T
e
x
t
i
l
e
6
9
0
6
0
A
D
1
2
8
1
(
1
2
9
5
)
1
3
8
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
1
2
2
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
6
3
2
T
e
x
t
i
l
e
6
9
0
6
0
A
D
1
2
8
1
(
1
2
9
5
)
1
3
8
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
4
8
0
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
1
-
6
4
1
M
u
s
c
l
e
6
9
5
7
5
A
D
1
2
7
6
(
1
2
9
3
)
1
3
8
9
F
o
c
a
c
c
i
1
9
8
2
:
7
4
A
Z
-
6
T
-
1
4
1
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
1
-
1
1
,
6
2
2
C
o
r
o
n
t
a
/
C
a
m
o
t
e
7
1
5
1
3
0
A
D
1
2
1
9
(
1
2
8
9
)
1
3
9
9
F
o
c
a
c
c
i
1
9
8
2
:
7
4
A
Z
-
1
4
0
T
-
1
6
M
a
i
t
a
s
-
C
h
r
i
b
a
y
a
B
e
t
a
-
7
8
9
6
6
T
e
x
t
i
l
e
7
4
0
6
0
A
D
1
2
5
1
(
1
2
8
4
)
1
2
9
8
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
2
4
B
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
9
6
4
T
e
x
t
i
l
e
7
4
0
6
0
A
D
1
2
5
1
(
1
2
8
4
)
1
2
9
8
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
5
6
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
6
2
9
T
e
x
t
i
l
e
7
4
0
7
0
A
D
1
2
4
0
(
1
2
8
4
)
1
3
0
0
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
X
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
1
,
6
2
1
S
o
r
o
n
a
7
6
5
7
5
A
D
1
2
1
9
(
1
2
7
8
)
1
2
9
4
F
o
c
a
c
c
i
1
9
8
2
:
7
4
A
Z
-
1
4
0
T
-
1
2
8
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
9
6
5
T
e
x
t
i
l
e
8
0
0
7
0
A
D
1
1
9
2
(
1
2
5
3
)
1
2
8
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
1
1
8
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
9
6
9
T
e
x
t
i
l
e
8
2
0
6
0
A
D
1
1
7
4
(
1
2
2
9
)
1
2
8
0
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
7
1
T
-
2
4
5
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
9
6
2
T
e
x
t
i
l
e
8
6
0
7
0
A
D
1
0
5
2
(
1
2
1
4
)
1
2
6
6
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
4
0
T
-
1
2
7
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
B
e
t
a
-
7
8
6
3
3
T
e
x
t
i
l
e
8
8
0
7
0
A
D
1
0
4
1
(
1
1
7
6
)
1
2
4
8
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
6
T
-
1
4
1
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
1
,
6
2
5
C
o
c
a
9
1
0
1
4
5
A
D
1
0
0
0
(
1
1
6
2
)
1
2
7
9
F
o
c
a
c
c
i
1
9
8
2
:
7
4
A
Z
-
1
4
0
T
-
1
0
5
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
1
-
1
2
-
3
3
6
M
u
s
c
l
e
9
8
0
8
0
A
D
9
9
6
(
1
0
2
8
)
1
1
6
5
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
4
0
T
-
2
3
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
2
-
3
4
9
M
u
s
c
l
e
9
8
0
8
0
A
D
9
9
6
(
1
0
2
8
)
1
1
6
5
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
4
0
T
-
3
6
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
1
-
1
2
-
3
3
7
M
u
s
c
l
e
1
0
0
0
8
0
A
D
9
8
4
(
1
0
2
2
)
1
1
5
9
M
u
o
z
1
9
8
3
:
8
3
P
L
M
-
9
T
-
2
4
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
1
-
6
2
4
S
o
r
o
n
a
1
0
0
8
8
0
A
D
9
7
8
(
1
0
2
0
)
1
1
5
6
F
o
c
a
c
c
i
1
9
8
2
:
7
4
A
Z
-
1
4
0
T
-
2
4
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
1
-
6
2
4
S
o
r
o
n
a
1
0
5
5
8
0
A
D
8
9
3
(
9
9
7
)
1
0
2
8
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
4
0
T
-
X
P
B
2
A
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
2
,
3
3
5
M
u
s
c
l
e
1
0
7
0
8
0
A
D
8
9
0
(
9
8
9
)
1
0
2
6
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
1
S
e
c
t
o
r
E
s
t
e
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
2
-
3
3
5
T
o
t
o
r
a
1
0
9
0
8
0
A
D
8
8
6
(
9
7
8
)
1
0
2
0
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
1
T
-
9
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
2
-
3
5
2
M
u
s
c
l
e
1
0
9
0
8
0
A
D
8
8
6
(
9
7
8
)
1
0
2
0
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
1
-
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
2
-
3
5
2
M
i
d
d
e
n
1
1
4
0
8
0
A
D
7
8
9
(
8
9
3
)
9
9
7
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
1
-
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
2
-
3
3
4
R
e
e
d
1
1
6
0
8
0
A
D
7
8
2
(
8
8
8
)
9
8
4
M
u
o
z
1
9
8
3
:
8
3
A
Z
-
1
4
0
T
-
7
5
M
a
i
t
a
s
-
C
h
i
r
i
b
a
y
a
I
-
1
2
-
3
4
8
M
u
s
c
l
e
1
1
9
0
8
0
A
D
7
2
7
(
8
8
1
)
9
6
7
M
u
o
z
1
9
8
3
:
8
3
S U T T E R 189
T
a
b
l
e
3
.
R
a
d
i
o
c
a
r
b
o
n
d
a
t
e
s
f
o
r
t
h
e
A
z
a
p
a
V
a
l
l
e
y
S
a
n
M
i
g
u
e
l
c
e
r
a
m
i
c
t
r
a
d
i
t
i
o
n
.
U
n
c
o
r
r
e
c
t
e
d
S
i
t
e
C
o
n
t
e
x
t
C
u
l
t
u
r
e
L
a
b
I
D
S
a
m
p
l
e
C
1
4
D
e
n
d
o
-
C
a
l
i
b
r
a
t
e
d
R
e
f
e
r
e
n
c
e
P
L
M
-
9
T
-
1
8
D
e
s
a
r
r
o
l
l
o
R
e
g
i
o
n
a
l
B
e
t
a
-
7
8
6
3
5
T
e
x
t
i
l
e
6
4
0
5
0
A
D
1
2
9
5
(
1
3
0
7
,
1
3
6
0
,
1
3
7
9
)
1
3
9
8
C
a
s
s
m
a
n
1
9
9
7
:
6
8
P
L
M
-
9
T
-
1
D
e
s
a
r
r
o
l
l
o
R
e
g
i
o
n
a
l
B
e
t
a
-
7
8
6
3
4
T
e
x
t
i
l
e
7
2
0
6
0
A
D
1
2
7
2
(
1
2
8
8
)
1
3
0
3
C
a
s
s
m
a
n
1
9
9
7
:
6
8
P
L
M
-
9
T
-
M
9
-
T
1
4
S
a
n
M
i
g
u
e
l
I
-
1
0
-
8
6
1
S
o
r
o
n
a
7
7
5
8
0
A
D
1
2
1
4
(
1
2
7
5
)
1
2
9
3
M
u
o
z
1
9
8
3
:
8
3
P
L
M
-
9
T
-
1
4
S
a
n
M
i
g
u
e
l
I
-
1
0
,
8
6
1
L
e
a
v
e
s
7
7
5
8
0
A
D
1
2
1
5
(
1
2
7
6
)
1
2
9
4
F
o
c
a
c
c
i
1
9
8
2
:
7
4
P
L
M
-
9
T
-
2
1
D
e
s
a
r
r
o
l
l
o
R
e
g
i
o
n
a
l
B
e
t
a
-
7
8
6
3
6
T
e
x
t
i
l
e
9
2
0
5
0
A
D
1
0
3
1
(
1
0
6
4
,
1
0
7
5
,
1
1
2
7
,
1
1
3
3
,
1
1
5
9
)
1
2
0
1
C
a
s
s
m
a
n
1
9
9
7
:
6
8
A
Z
-
1
1
T
-
8
S
a
n
M
i
g
u
e
l
-
M
u
s
c
l
e
9
7
0
8
0
A
D
1
0
0
4
(
1
0
3
2
)
1
1
6
8
M
u
o
z
1
9
8
3
:
8
3
F
i
g
u
r
e
2
:
A
z
a
p
a
V
a
l
l
e
y
a
r
c
h
a
e
o
l
o
g
i
c
a
l
s
i
t
e
s
d
i
s
c
u
s
s
e
d
b
y
t
h
i
s
s
t
u
d
y
.
190 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
The cemetery occupies the southern slope of the
Azapa Valley directly across from Late Intermedi-
ate period sites Azapa-6 and Azapa-71. Burials were
placed in sandy cavities in a seated position. Grave
lots reect a mixed agropastoral and maritime sub-
sistence. Of the seventy-seven remains examined
from Azapa-140, sixty-nine were associated with
Maitas and Maitas-Chiribaya ceramics, while only
four were associated with San Miguel ceramics.
The implications of associated ceramics will be
examined in greater detail in subsequent sections
of this paper.
3) Azapa-71. Azapa-71 (AZ-71) is primarily a Late
Intermediate period cemetery located 12.5 km
from the coast of Arica along the northern slope
of the Azapa Valley (Espinoza et al. 1994; Focacci
1961, 1968, 1993). This cemetery is located
approximately 150 m west of the contemporane-
ous Late Intermediate period cemetery Azapa-6
described below. Associated grave offerings reect
that agropastoral activities were practiced by indi-
viduals interred at the site. Tombs from Azapa-71
contained single, exed burials wrapped in one
or two wool shirts of relatively unremarkable
quality. Burials were placed in sandy pit tombs
in a seated exed position (Focacci 1993). The
overwhelming majority of burials at Azapa-71
were associated with Cabuza style ceramics. Only
two of the forty-three Late Intermediate period
remains from Azapa-71 examined by this study
were associated with non-Cabuza ceramics. Once
again, this information will be examined in greater
detail below.
4) Azapa-6. Azapa-6 (AZ-6) is a Late Intermediate
period cemetery located 13 km from the coast of
Arica along the northern slope of the Azapa Val-
ley (Espinoza et al. 1994; Focacci 1961, 1993).
This cemetery is located adjacent to the modern
cemetery of the town of San Miguel and the con-
temporaneous Late Intermediate period cemetery
Azapa-71 described above. Mortuary practices for
Azapa-6 are nearly identical to those described
above for Azapa-71 and will not be discussed fur-
ther. The majority of Azapa-6 remains examined
in this study were associated with Cabuza style
ceramics. Of the forty remains from Azapa-6
examined by this study, twelve were associated
with non-Cabuza Late Intermediate period
ceramics. These tombs were associated with
San Miguel (n = 7) and Maitas-Chiribaya (n = 5),
and were spatially separated from Cabuza tombs
within the cemetery (Focacci 1993; Muoz and
Focacci 1985). As already mentioned for the other
three sites examined by this study, these data will
be examined in greater detail below.
II. Genetic Relations
Genetic relatedness may be considered analogous to
unconscious (i.e., isochrestic) stylistic variation that is
related to group identity, because it also represents an
unintended consequence of the degree of reproduc-
tive relations among groups. While, at some level,
we always expect to encounter some degree of genetic
difference among groups even those with recent
common ancestorsbiodistance measures provide
valuable information regarding the extent to which
active expressions of group afliation and distinction
are related to genetic difference.
Genetic relations among the mortuary populations
examined by this study were estimated using both dental
and cranial epigenetic traits. Dental traits were scored
using standardized descriptions and casts (Turner,
Nichol, and Scott 1991), while cranial traits were
scored using a standardized list of traits and descrip-
tions (Buikstra and Ubelaker 1994). Once collected, the
frequencies for each set of epigenetic traits were arcsine
transformed before calculating the standardized mean
measure of divergence (Berry and Berry 1967; Green
and Myers Suchey 1976; Lukacs and Hemphill 1991;
Sjvold 1973). The matrices of standardized mean
measures of divergence for both dental and cranial
epigenetic traits were analyzed using hierarchical clus-
tering procedures.
III. Cranial Deformations
Cranial deformations represent an intentional (i.e.,
emblemic), culturally inuenced practice of altering
skull shape. Skull form is usually altered through appli-
cation of pressure to the skull using pads and boards.
The process usually begins during infancy (Guilln
1992; Hoshower et al. 1995; Lozada, Blom, and Buiks-
tra 1996). Deformations are thought to be particularly
promising in the identication of prehistoric ethnicity,
given the ethnohistoric evidence indicating that cranial
deformation was used by Andeans to express ethnic dif-
ferences (Cobo 1983 [1653]), and the fact that cranial
deformations represent a permanent, culturally medi-
ated behavior that can be directly associated with the
ancient people in question.
S U T T E R 191
While a number of studies have proposed different
methods for classifying and describing cranial defor-
mations (Hoshower et al. 1995; Lozada, Blom, and
Buikstra 1996; also see Blom and Lozada and Buikstra
chapters, this volume), no clear standard currently
exists. I rely upon widely recognized descriptions that
are largely based upon skull form. Cranial deformations
encountered during this investigation include normal or
undeformed skulls, tabular erect, tabular oblique, annu-
lar (also referred to as turban), and fronto-occipital
round. While it is recognized that this scheme does not
provide detailed information on the materials used and
location of pressure used to deform the skull, I chose it
for the purposes of replicability and simplicity.
IV. Dental Pathologies
In order to evaluate ethnic differences that might be
reflected in dietary differences, dental pathologies
were examined for human remains that were dentally
mature (i.e., have a full complement of permanent den-
tition). The dental pathologies examined for this study
include caries (or cavities) and premortem tooth loss.
Dental caries are a disease process that result from the
demineralization of a tooths enamel surface by acids
created by bacteria (Lactobacillus acidophilus, Streptococcus
mutans, and many others). These bacteria inhabit the
plaque that covers tooth surfaces and ferment dietary
sugars (Menaker 1980; Newburn 1978, 1982). Dental
caries and tooth loss are often used to infer subsistence
and the relative proportions of carbohydrates among
prehistoric peoples (Moore and Corbett 1971; Powell
1985; Turner 1979).
A number of factors contribute to an individuals car-
ies susceptibility. Dietary carbohydrates are the only
substances that cariogenic bacteria consumefats
and proteins do not contribute to caries. Therefore,
those individuals who consume great quantities of car-
bohydrates will be at greater risk of obtaining caries.
Vigorous mastication increases the natural cleansing
of the mouth by increasing salivary ow, producing
an inhibitory effect on caries. Dental wear effectively
decreases the surface area of a tooth that is exposed
to carious activities (Powell 1985). The age of an
individual inuences their susceptibility to caries and
dental disease; older individuals have teeth that have
been exposed to carious processes for longer periods of
time than the teeth of younger individuals.
Dental pathologies related to diet will, at times, be
analogous to isochrestic variation identied among
material objects (Sackett 1982), in that, on a group
level, these pathologies may discriminate among
ethnic groups when differences in the levels of dental
pathologies represent a consequence of group-specic
subsistence activities.
For this study, tooth surfaces and sockets were visu-
ally inspected for caries and premortem tooth loss, and
the Diseased-Missing Index was calculated for each indi-
vidual (Moore and Corbett 1971). This index uses the
ratio of carious teeth and teeth missing prior to death
relative to the total number of teeth and toothless sock-
ets observed for each individuals dentition. An implicit
assumption made when using this index is that premor-
tem tooth loss results from carious activities. There are
certainly other factors that may cause premortem tooth
loss. However, there is substantial clinical evidence sup-
porting the use of this index as an indication of carious
activities (Harris 1968; Menaker 1980).
For comparisons of the Diseased-Missing Index,
dental wear and age-at-death were used as covariates
during Multivariate Analysis of Covariance (MAN-
COVA) statistical procedures. This procedure permits
control for differences in age and wear so that the rela-
tionship among carious indicators and the independent
variables can be investigated (Sutter 1995). Independent
variables examined by this study include site, ceramic
association, and cranial deformation. The interactions
among these variables was also explored. A 0.05 level
of signicance was used for all analyses.
RESULTS
I. Grave Good Analysis
1) Playa Miller-4. While the attempt to completely
register tombs and grave goods from Playa Miller-
4 is currently in progress, nearly all burials were
accompanied by Late Intermediate period San
Miguel style ceramics. A small percentage of
the tombs were interred with Maitas-Chiribaya
ceramics. None of the burials from Playa Miller-4
contained Cabuza style ceramics. Grave offerings
clearly reect a maritime subsistencemale buri-
als contained shing tackle, harpoons, chopes (tools
made of modied sea lion ribs used for prying
off and opening shell sh), and model boats, and,
occasionally, males were interred with copper
sh hooks. Female remains from Playa Miller-4
were often interred with textiles tools, wool, and
dye boxes. Nearly all burials of both sexes and all
ages were interred with food offerings of dried
192 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
sh, mollusks, as well as some agricultural prod-
ucts (primarily maize). Playa Miller-4 grave lots
contained the richest and highest quality grave
offerings of all the Late Intermediate period
Azapa Valley sites examined by this study (Cass-
man 1997).
2) Azapa-140. The grave lots of Azapa-140 burials
include items that suggest a mixed agropastoral
and maritime subsistence. Much like contempo-
raneous burials from Playa Miller, males were
often accompanied by shing gear and model
boats, but were also often interred with digging
sticks, while weaving toolkits and materials for the
production of textiles were often included with
the grave offerings of females. Azapa-140 graves
commonly contained agricultural food offerings,
such as maize, porotos (beans), aji peppers, dried
sh, and a pair of llama feet.
Analysis of the grave offerings indicates that indi-
viduals interred at Azapa-140 were accompanied by
fewer and poorer quality offerings than those recorded
from other sites examined by this study (Espinoza et al.
1994). Cassman (1997) reports that graves from Azapa-
140 were wrapped in one or two wool shirts of relatively
poor quality when compared with those from Azapa-71
and contemporaneous Late Intermediate period Playa
Miller sites.
As noted above, of the seventy-seven remains exam-
ined from Azapa-140, sixty-nine were associated with
Maitas and Maitas-Chiribaya ceramics, while only four
were associated with San Miguel ceramics. None of the
burials from this cemetery were associated with ceramics
of the Cabuza tradition. Interestingly, of the four burials
associated with San Miguel ceramics, three were adult
females, while one was that of a subadult male twelve
years of age. To conrm this pattern, I also examined
information on the graves and grave goods that did not
qualify for this study (i.e., juvenile remains, those miss-
ing skulls, and unwrapped fardos, or mummy bundles).
Of the 126 burials registered for Azapa-140, only eight
contained nonMaitas-Chiribaya (i.e., San Miguel)
ceramics. Of those burials, four were adult females and
the remaining four were subadults (i.e., <12) or infants
that were often interred with the females.
3) Azapa-71. Grave lots from Azapa-71 reflect
agropastoral economic activities. Grave offer-
ings included ceramics, agricultural tools, a pair
of llama feet, and agricultural offerings, such as
maize, porotos, coca, and aji peppers. Relatively
little intrasite social stratification was evident
among the grave lots from Azapa-71. When
wealthier burials did occur, they tended to be
tombs that included textiles of higher quality
(Cassman 1997) and ceramics of more than one
ceramic tradition (i.e., both Cabuza and San
Miguel ceramics). According to Cassman (1997),
burials at Azapa-71 were wrapped in one or two
wool shirts of relatively unremarkable quality,
slightly better than AZ-140, but clearly of lower
quality than those associated with coastal Playa
Miller burials.
As mentioned above, the overwhelming majority of
burials at Azapa-71 were associated with Cabuza style
ceramics. Only two of the forty-three Late Intermedi-
ate period remains from Azapa-71 examined by this
study were associated with non-Cabuza ceramics. Both
of these burials were those of females associated with
Maitas-Chiribaya ceramics. To conrm this pattern,
I also examined information on the graves and grave
goods that did not qualify for this study (i.e., juvenile
remains, those missing skulls, and unwrapped fardos).
Of the ninety-eight Late Intermediate period burials
registered for Azapa-71, three (two adult females and
one juvenile) were associated with San Miguel ceram-
ics, while only four burials (three adult females and the
remains of a fetus) were associated with Maitas-Chiri-
baya ceramics.
4) Azapa-6. Interestingly, with the exception of three
tombs of adult males (one associated with Maitas-
Chiribaya ceramics, the other two associated with
San Miguel ceramics), those tombs containing
non-Cabuza ceramics were those of females and
subadults. To conrm this pattern, I also exam-
ined information on the graves and grave goods
that did not qualify for this study. Of the fty-
eight Late Intermediate period tombs registered
for Azapa-6, thirty-nine were associated with
Cabuza ceramics, while nineteen were interred
with non-Cabuza (i.e., either Maitas-Chiribaya
and San Miguel) ceramics.
While the patterns for Azapa-6 are not as universal as
those described above for the other sites examined here,
nearly all tombs associated with non-Cabuza ceramics
were those of females and subadults (n = 15). Indeed, of
the ten burials interred with Maitas-Chiribaya ceram-
ics, only one was that of an adult male, and only three
of the nine burials interred with San Miguel ceramics
were those of adult males.
S U T T E R 193
II. Genetic RelatednessEpigenetic Trait Analysis
In examination of epigenetic relations as evaluated using
dental traits, I have previously reported (Sutter 1994,
1996, 1997, 2000) that biodistance measures for the
Azapa Valley are small and insignicant (gure 3). A
closer examination of these data indicates that, although
relations among Azapa Valley mortuary populations is
spatially insignicant (table 4), there are some noteworthy
patterns. More specically, the dental trait data indicate
that there is some degree of genetic continuity among the
Late Intermediate period population from Playa Miller-4
and earlier El Laucho and Chinchorro coastal populations
(gure 4). Indeed, among the Late Intermediate period
mortuary populations examined by this study, the popu-
lation from Playa Miller-4 is the least similar to other
contemporaneous coastal valley populations.
Figure 3: Multidimensional scaling solution for 12 south-central Andean samples.
Table 4. Standardized Mean Measure of Distance (stMMD) values based upon non-metric dental traits for eight prehistoric Azapa Valley mortuary
populations.
Alto Ramirez Azapa - 6 Azapa - 71 Azapa - 8 Playa Miller - 4 Playa Miller -7 Azapa - 140 Chinchorro
Alto Ramirez . . . . . . . .
Azapa - 6 -0.400 . . . . . . .
Azapa - 71 -0.256 -0.182 . . . . . .
Azapa - 8 -0.155 0.091 0.004 . . . . .
Playa Miller - 4 -0.013 0.283 0.243 -0.334 . . . .
Playa Miller -7 -0.391 -0.284 -0.313 -0.247 -0.290 . . .
Azapa - 140 -0.075 -0.005 -0.195 0.174 0.299 -0.386 . .
Chinchorro -0.142 0.497 0.078 0.020 0.119 -0.272 -0.314 .
194 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Figure 4: Cluster analysis of standardized Mean Measure of Distance (stMMD) values calculated using non-metric dental traits for Azapa Valley
mortuary populations.
Figure 5: Cluster analysis of standardized Mean Measure of Distance (stMMD) values calculated using non-metric cranial traits for Azapa Valley
mortuary populations.
S U T T E R 195
Using epigenetic cranial traits to estimate genetic
relations among the same Azapa Valley populations,
Sutter and Mertz (2004) report similar biodistance
results to those obtained here using non-metric dental
traits. Analysis of the cranial data indicates small and
non-signicant biological distances (table 5), indicat-
ing genetic continuity with relatively unsubstantial
gene flow among prehistoric Azapa coastal valley
populations. As reported for results obtained using
dental traits, it is apparent that the mortuary population
from Playa Miller-4 is least similar to contemporaneous
coastal valley sites, and more similar to earlier coastal
populations (gure 5).
III. Cranial Deformations
Results of the
2
analysis of cranial deformation by
site indicate that there are no statistically signicant
associations in these data (table 6). With the exception
of the tabular oblique deformation at Azapa-71, there
is no deformation that is found at frequencies greater
than 50% at any given site. Indeed, all deformations
are found, albeit in varying frequencies, at all four of
the sites examined by this study. The annular deforma-
tion is found at its highest frequency at the coastal site,
Playa Miller-4, whereas a lack of cranial deformations
predominates at Azapa-140. Azapa-6 is characterized
by relatively high frequencies of the tabular oblique
deformation and the absence of cranial deformations
(i.e., normal). A possible, albeit weak, pattern in these
data may be the predominance of tabular oblique
deformations at the Cabuza sites, Azapa-6 and Azapa-
71. Possibly a second pattern, although difficult to
evaluate given the lack of comparative coastal sites, is
the predominance of the annular deformation among
crania from Playa Miller-4.
Chi-squared analysis of cranial deformation by cul-
ture (i.e., ceramic association) also failed to produce
statistically signicant results (table 7). Likewise, all
deformations are associated among all three of the
ceramic traditions examined by this study. There was
a predominance of the tabular oblique deformation
Table 5. Standardized Mean Measure of Distance (stMMD) values based upon non-metric cranial traits for eight prehistoric Azapa Valley mortuary
populations.
Alto Ramirez Azapa - 6 Azapa - 71 Azapa - 8 Playa Miller - 4 Playa Miller -7 Azapa - 140 Chinchorro
Alto Ramirez . . . . . . . .
Azapa - 6 -0.061 . . . . . . .
Azapa - 71 -0.280 -0.317 . . . . . .
Azapa - 8 -0.308 -0.248 -0.403 . . . . .
Playa Miller - 4 0.283 0.309 0.037 -0.078 . . . .
Playa Miller -7 0.443 0.494 0.211 0.338 0.050 . . .
Azapa - 140 -0.161 -0.228 -0.498 -0.297 0.514 0.683 . .
Chinchorro 0.077 -0.060 -0.351 -0.431 0.029 0.419 -0.150 .
Table 6. Cranial deformation by site.
Deformation Type Playa Miller - 4 Azapa - 6 Azapa - 71 Azapa - 140 Overall
Normal 23.7% (9) 32.3% (10) 19.4% (7) 35.5% (22) 28.7%
Tabular Erect 5.3% (2) 3.2% (1) 13.9% (5) 3.2% (2) 6.0%
Tabular Oblique 7.9% (3) 32.3% (10) 50.0% (18) 27.4% (17) 28.7%
Annular 44.7% (17) 6.5% (2) 13.9% (5) 24.2% (15) 23.4%
Fronto-Occipital Rounding 18.4% (7) 25.8% (8) 2.8% (1) 9.7% (6) 13.2%
TOTAL 38 31 36 62 167
*Values in italics represent the most common form of cranial deformation for each column.
196 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
among burials associated with Cabuza ceramics,
whereas the annular deformation tends to charac-
terize burials associated with San Miguel ceramics.
Burials interred with ceramics of the Maitas-Chiri-
baya tradition were characterized by slightly higher
frequencies of non-deformed skulls than of tabular
oblique skulls.
IV. Dental Pathologies
None of the statistical analyses of the DM Index by
sex produced statistically signicant results, so sex
was dropped as a variable from subsequent statistical
analyses. MANCOVA analysis of site by cranial defor-
mation also produced non-signicant results (table 9).
Indeed, examination of these data clearly indicates the
lack of any clear pattern (gure 6). However, when
these data were examined by site alone, statistically
signicant patterns emerged (gure 7). Specically,
the maritime population represented by the mortu-
ary sample from Playa Miller-4 exhibits the lowest
levels of dental pathologies, while the mixed mari-
time-agropastoral mortuary sample from Azapa-140
also exhibits lower frequencies of dental pathologies
when compared to the agropastoral Cabuza mortuary
samples from Azapa-6 and Azapa-71. The levels of
dental pathologies at Playa Miller-4 and Azapa-140
were signicantly lower than those observed for the
Azapa-6 and Azapa-71 mortuary samples.
Results of MANCOVA analyses of the DM Index
produced statistically signicant results when dental
pathologies were examined by both culture (i.e., ceramic
association) and deformation; however, the interaction
among these variables was not signicant (table 8).
When examined by culture, the DM Index is clearly
found at higher levels among the dentitions of individu-
als associated with Cabuza ceramics (gure 8).
Finally, when the dental pathologies are examined
by cranial deformation (gure 9), it is apparent that
individuals exhibiting a lack of cranial deformations
also exhibit signicantly fewer dental pathologies than
individuals exhibiting the tabular oblique, annular, or
fronto-occipital round deformation. Although indi-
viduals exhibiting the tabular erect deformation also
exhibit relatively fewer dental pathologies than indi-
viduals exhibiting other cranial deformations, too few
individuals exhibited the tabular erect deformation for
this difference to be signicant.
DISCUSSION AND CONCLUSIONS
Based upon the results presented here I tentatively
suggest that ethnicity among Late Intermediate period
Azapa Valley mortuary populations is, to varying
degrees, evident in all the indicators examined by this
study. In contrast to studies presented by Blom, and
Lozada and Buikstra in this volume, cranial deforma-
tions were the least useful among all the classes of data
examined by this investigation. While some patterns
may exist for some cranial deformations (i.e., possibly
tabular oblique and annular deformations), the mor-
tuary populations examined by this study are largely
characterized by a lack of ethnically based variation in
cranial deformations.
Each of the cranial deformations examined by
this study was present at all of the sites. The most
conspicuous among the cranial deformations was the
tabular oblique. Although not statistically signicant,
the tabular oblique deformation was most often asso-
ciated with crania from the agropastoral Cabuza sites
Azapa-6 and Azapa-71.
The annular deformation may also be weakly asso-
ciated with ethnicity. This deformation was found at
its highest frequencies among crania from the coastal
site Playa Miller-4. While it is difcult to attribute any
signicance to this nding, given the lack of any clear
patterns among the mortuary samples examined here,
other studies have also reported a predominance of this
cranial deformation among coastal sites (Lozada, Blom,
and Buikstra 1996). It is possible that this deformation
is associated with ethnically coastal peoples.
When the associations among cranial deformations
and dental pathologies are explored, these results also
failed to exhibit any clear pattern, save one comparison.
The only individuals exhibiting a signicant difference
from the others in relation to deformations and dental
pathologies were those characterized by non-deformed
skulls (i.e., those with normal crania). A cursory exami-
nation of grave good data for those individuals failing
to exhibit cranial deformations suggests that this dif-
ference may be a class-based association, rather than
an ethnicity-based one. Individuals with non-deformed
crania were generally interred with fewer grave goods
(Espinoza et al. 1994) and textiles of poorer quality
(Cassman, personal communication 1995).
While these results tenuously suggest that cranial
deformations may have been used by coastal and Cabuza
populations to express group afliation, they may also
have been used to indicate class or some other, as of
yet unidentied, relationship among inland populations.
S U T T E R 197
Table 7. Cranial deformation by culture (ceramic association).
Deformation Type Maitas Chiribaya San Miguel Cabuza Overall
Normal 34.4% (22) 25.5% (12) 25.0% (14) 28.7%
Tabular Erect 3.1% (2) 4.3% (2) 10.7% (6) 6.0%
Tabular Oblique 29.7% (19) 18.5% (5) 42.9% (24) 28.7%
Annular 20.3% (13) 40.4% (19) 12.5% (7) 23.4%
Fronto-Occipital Rounding 12.5% (8) 19.1% (9) 8.9% (5) 13.2%
TOTAL 64 47 56 167
*Values in italics represent the most common form of cranial deformation for each column.
Table 8. Diseased-Missing Index MANCOVA Analysis: Culture by Deformation.
Source Sum-of-Squares DF Mean-Square F-Ratio p - Value
Culture 918.57 2 459.29 3.22 0.04
Deformation 1805.89 4 451.47 3.16 0.02
Culture* Deformation 324.21 8 40.53 0.28 0.97
Covariates
Age 822.76 1 822.76 5.76 0.02
Wear 178.29 1 178.29 1.25 0.27
Error 14989.5 105 142.76
*Values in italics are statistically signicant at the 0.05 level.
Table 9. Diseased-Missing Index MANCOVA Analysis: Site by Deformation.
Source Sum-of-Squares DF Mean-Square F-Ratio p - Value
Site 293.94 3 97.81 0.69 0.56
Deformation 1545.11 4 386.28 2.71 0.03
Site* Deformation 1586.73 12 132.23 0.93 0.52
Covariates
Age 1165.13 1 1165.13 8.16 0.01
Wear 79.22 1 79.22 0.56 0.46
Error 15706.39 110 142.79
*Values in italics are statistically signicant at the 0.05 level.
198 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
The lack of any clear, statistically signicant patterns
among crania from the coastal Azapa Valley makes any
claims regarding the deformations difcult to make.
These relationships will require more detailed analyses
of deformation practices and grave goods than what has
been presented here.
It does not appear that textilesa potentially highly
visible, culturally mediated indicator of ethnicitywere
being used to express ethnic distinctions among Late
Intermediate period Azapa Valley populations. Cass-
man (1997) reports that textiles associated with Late
Intermediate period burials from Azapa-140, Azapa-71,
and Playa Miller shared identical motifs, colors, and
manufacturing techniques. The primary differences
among these sites were differences in number and
quality of associated textiles. Late Intermediate period
burials from Playa Miller were interred with more and
the highest quality textiles, relative to the other two
sites examined by Cassman. Burials from Azapa-71 and
Azapa-140 were interred with textiles of lower quality,
with textiles from Azapa-71 being only slightly better in
quality than those associated with Azapa-140 burials.
The most apparent indications of ethnicity among
Late Intermediate period Azapa Valley populations are
derived when the associations among ceramic style,
biodistance analyses, and shared economic activities as
inferred using dental pathologies and grave furniture
are considered in light of one another. Both dental and
Figure 6: Diseased and missing index (DM Index) by both site and cranial deformation.
S U T T E R 199
F
i
g
u
r
e
7
:
D
i
s
e
a
s
e
d
a
n
d
m
i
s
s
i
n
g
i
n
d
e
x
(
D
M
I
n
d
e
x
)
b
y
s
i
t
e
o
n
l
y
.
200 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
cranial traits indicate that longstanding, albeit statisti-
cally insignicant, genetic relations existed among the
coastal populations from Playa Miller-4 and earlier
coastal people of the Chinchorro and El Laucho (i.e.,
Playa Miller-7) traditions. Further, the Playa Miller-4
mortuary sample is least similar to contemporaneous
samples from the Azapa Valley interior. Conversely,
both dental and cranial traits indicate that longstand-
ing, though statistically insignicant, genetic relations
existed among the inland populations from Azapa-6,
Azapa-71, and Azapa-140 and earlier agropastoral
populations represented by the Formative period Alto
Ramrez tradition.
Grave lots and dental pathologies from Playa Miller-
4 clearly indicate that this population was characterized
by a maritime subsistence. Furthermore, individuals
interred at this site were almost exclusively associated
with San Miguel ceramics. Grave lots and dental pathol-
ogies for the mortuary population at Azapa-140 clearly
reect a mixed maritime-agropastoral economy: both
agricultural and maritime implements and food offer-
ings were interred as grave offerings in the tombs of
individuals interred at the site. Late Intermediate period
burials from Azapa-6 and Azapa-71 are clearly associ-
ated with both implements and food offerings reecting
that individuals interred at these sites participated in
Figure 8: Diseased and missing index (DM Index) by culture (ceramic association) only.
S U T T E R 201
agropastoral activities. The signicantly higher inci-
dence of dental pathologies among burials from these
sites conrms that individuals at Azapa-6 and Azapa-71
consumed greater quantities of carbohydrates obtained
through agricultural goods than did their contempo-
raries at Playa Miller-4 and Azapa-140.
While a cursory analysis of the associated ceramics
from these cemetery sites does not appear to conform to
the patterns observed among the genetic and economic
activities that characterize each of these sites, a closer
inspection reveals that they do. As previously men-
tioned in the results section, in the few instances when
burials from Azapa-140a cemetery overwhelmingly
characterized by Maitas-Chiribaya ceramicswere
interred with San Miguel ceramics (n = 8), those buri-
als were always those of adult females and juveniles or
infants. Similarly, at Azapa-71a cemetery character-
ized by an abundance of Late Intermediate period
Cabuza ceramicsthe few burials (n = 6) interred
with either Maitas-Chiribaya or San Miguel ceram-
ics were always those of adult females and juveniles
or infants. Although this pattern was not as universal
at the predominantly Cabuza site Azapa-6, nearly all
(with the exception of four) burials from this cemetery
that contained non-Cabuza ceramics were also those
of adult females and juveniles. Unfortunately, a closer
Figure 9: Diseased and missing index (DM Index) by cranial deformation only.
202 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
inspection of those Playa Miller-4 burials associated
with Maitas-Chiribaya ceramics will not be possible
until the ongoing registry of grave goods is complete.
The patterns in the ceramics associations among
each of the sites examined by this study strongly suggest
that those burials associated with non-local ceramics
represent women and their children who were pres-
ent at these sites as a result of exogamous marriages.
Interestingly, Cabuza ceramics were not associated with
any of the burials from Playa Miller-4 or Azapa-140.
Indeed, despite years of extensive site surveys of the
Azapa Valley, Cabuza ceramics have never been found
in association with either coastal or predominantly Mai-
tas-Chiribaya sites (Muoz 1983). Another tantalizing
and related aside is that preliminary analyses of dental
data in my possession
1
clearly indicate that mortuary
populations at Azapa-6, Azapa-71, and Azapa-140 were
characterized by patrilocal residence, whereas those
interred at Playa Miller-4 were characterized by either
matrilocal or avunculocal residence.
When considered in light of this information,
Late Intermediate period ceramic styles from the
Azapa Valley do appear to signify ethnic afliations.
More specically, it is likely that Cabuza ceramics are
associated with agropastoral populations, whereas Mai-
tas-Chiribaya and San Miguel ceramics are associated
with mixed maritime-agropastoralists and coastal sh-
ing villages, respectively. These characterizations will
require further investigation, particularly as they relate
to the incompletely analyzed Playa Miller-4 site.
The patterns among ceramics, genetic relatedness,
and shared economic activities provide tantalizing
evidence regarding the factors underlying Late Interme-
diate period ethnicity in the Azapa Valley, Chile. Results
from this investigation suggest to me that associations
among the ceramics and the results of epigenetic analy-
ses represent two different forms of ethnic variation,
with ceramic traditions representing an overt expression
of ethnic afliation, while the genetic relatedness repre-
sents an unintentional byproduct of cultural behaviors
(i.e., marriage and residence patterns).
As to why these ethnic differences may have
occurred, I am compelled to argue that among Late
Intermediate period Azapa Valley populations, both
genetic afnities and actively expressed group aflia-
tion are the result of what Carter Bentley (1987:35) has
described as situationally shared elements of a multi-
dimensional habitus. In other words, the underlying
cause of ethnicity among Late Intermediate period
Azapa Valley groups is derived from commonly shared
economic interests and practices. For these coastal val-
ley populations, it was through practice that common
perceptions, experiences, and world view emerged.
Subsequently, an awareness of group/other evolved
to become what is archaeologically perceived to be
ethnic differences.
Some important lessons can be derived from the
results of this investigation. In this case, culturally based
expressions of group afliation among these mortuary
populations appear to have been ceramics and eco-
nomically related grave furniture and food offerings.
Apparently, among a constellation of possible culturally
based practices and material objects (i.e., textiles, ceram-
ics, cranial deformations, etc.), it appears that ceramics
represent the only unambiguous consciously chosen cor-
relate (i.e., emblemic style) of ethnicity. It is important
to point out that this association is arbitrary and may
change from region to region or chronological period
to chronological period. The signicance of differences
among economically related grave furniture and food
offerings likely represents an unconscious reection
of group afliation that was derived from commonly
shared activities (i.e., isochrestic variation).
Finally, ethnic differences among Late Intermediate
period Azapa Valley populations were not the result of
the arrival of nonlocal ethnic colonists, nor the result
of political group distinctionsa conclusion that other-
wise might not have been reached using an exclusively
archaeologically based methodology (Sutter 2000).
Conversely, ethnic differences (i.e., actively expressed
group afliations and distinctions) clearly did exist in
the Azapa Valley during this perioda conclusion that
might not otherwise have been reached using a strictly
biologically based methodology.
Within this paper it has been my goal to identify
both the material correlates and sources for the for-
mation of ethnic groups using a bioarchaeological
approach. Hopefully, this methodology will help future
investigators to discriminate when observed variation
in material culture and culturally based practices is the
result of group identity derived from commonly shared
group activities, or the result of the presence of geneti-
cally unrelated populations, such as ethnic colonists.
NOTES
1. I am currently working on an unambiguous method of
evaluating residence patterns using dental traits.
S U T T E R 203
REFERENCES
Aldenderfer, M. S., and C. Stanish
1993 Domestic architecture, household archaeology,
and the past in the south-central Andes. In Domes-
tic architecture, ethnicity, and complementarity in the
south-central Andes, edited by M. S. Aldenderfer,
112. Iowa City: University of Iowa Press.
Barth, F.
1969 Introduction. In Ethnic groups and boundaries: The
social organization of culture difference, edited by F.
Barth, 938. Boston: Little, Brown and Company.
Bawden, G.
1993 An archaeological study of social structure and
eth nic replacement in residential architecture of the
Tumilaca Valley. In Domestic architecture, ethnicity,
and complementarity in the south-central Andes, edited
by M. S. Aldenderfer, 4254. Iowa City: University
of Iowa Press.
Bermann, M.
1994 Lukurmata: Household archaeology in prehispanic
Bolivia. Princeton, NJ: Princeton University Press.
Berreman, G. D.
1972 Social categories and social interaction in urban
India. American Anthropologist 73 (3): 567586.
Berry, A. C., and R. J. Berry
1967 Epigenetic variation in the human cranium. Journal
of Anatomy 101:361379.
Buikstra, J. E.
1995 Tombs for the living . . . or . . . for the dead: The
Osmore ancestors. In Tombs for the living: Andean
mortuary practices, edited by T. Dillehay, 229280.
Washington, D.C.: Dumbarton Oaks Research
Library and Collection.
Buikstra, J. E., and D. H. Ubelaker, eds.
1994 Standards for data collection from human skeletal
remains. Arkansas Archaeological Survey Research
Series 44. Fayetteville.
Carr, C., and J. Neitzel
1995 Future directions for material style studies. In Style,
society, and person: Archaeological and ethnological per-
spectives, edited by C. Carr and J. Neitzel, 437459.
New York: Plenum Press.
Carter Bentley, G.
1987 Ethnicity and practice. Comparative Studies in Society
and History 29:2455.
Cassman, V.
1997 A reconsideration of prehistoric ethnicity and sta-
tus in northern Chile: The textile evidence. Ph.D.
dissertation, Department of Anthropology, Arizona
State University.
Cobo, B.
1983 [1653] History of the Inca Empire. Translated by
R. Hamilton. Austin: University of Texas Press.
Cohen, A.
1981 The politics of elite culture: Explorations in the drama-
turgy of power in a modern African society. Berkeley:
University of California Press.
Drummond, L.
1980 The cultural continuum: A theory of intersystems.
Man 16 (4): 352374.
Espinoza, G., T. Caipa, L. Latorre, V. Cassman, and R.
Sutter
1994 Registro computacional de las colecciones arqueolgicas
Azapa 6, Azapa 71a, Azapa 71b, Azapa 140, y Azapa
141. Arica, Chile: Departamento de Arqueologa y
Museologa, Universidad de Tarapac.
Espoueys, O., V. Schiappacasse, M. Uribe, and J. Berenguer
1994 En torno al surgimiento de la cultura Arica. Paper
presented at XIII Congreso Nacional de Arque-
ologa Chilena, Antofogasta.
Focacci, G.
1961 Excavaciones en San Miguel de Azapa. Boletnes del
Museo Regional de Arica 6.
1968 Boletas de campo: Excavaciones en AZ-71. Notes
on le, Museo San Miguel de Azapa, Universidad
de Tarapac, Arica.
1969a Arqueologa de Arica, secuencia del perodo agro
algarero, Horizonte Tiahuanacoide. In Actas del V
Congreso de Arqueologa Chilena, edited by M. N. d.
H. Natural, 2125. Santiago: Sociedad de Arque-
ologa Chilena.
1969b Boletas de campo: Excavaciones en Playa Miller 4,
2, y 6. Notes on le, Museo San Miguel de Azapa,
Universidad de Tarapac, Arica.
1974 Excavaciones en Playa Miller 7. Chungar 3:2374.
1980 Sntesis de la arqueologa del extremo norte de
Chile. Chungar 6:323.
1982 Excavaciones en el cementerio de Playa Miller 9. Documen tos
de Trabajo 2: 126213. Arica: Instituto de Antropologa
y Arqueologa, Universidad de Tarapac.
1993 Excavaciones arqueolgicas en el cementerio AZ-6
Valle de Azapa. Chungar 24/25:69124.
Geary, P.
1983 Ethnic identity as a situational construct in the early
Middle Ages. Mitteillungen der Anthropologischen
Gesellschaft in Wien 113:1526.
Goldstein, P.
1993 Tiwanaku temples and state expansion: A Tiwan-
aku sunken-court temple in Moquegua, Peru. Latin
American Antiquity 4 (1): 2247.
204 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
1995 Tiwanaku settlement patterns of the Azapa Valley,
Chile: New data and the legacy of Percy Dauels-
berg. Dilogo Andino 14/15:5774.
Green, R. F., and J. Myers Suchey
1976 The use of inverse sine transformations in the analy-
sis of non-metric cranial data. American Journal of
Physical Anthropology 45:6168.
Guilln, S. E.
1992 The Chinchorro Culture: Mummies and crania in the
reconstruction of preceramic coastal adaptation in the
south-central Andes. Ph.D. dissertation, Department
of Anthropology, University of Michigan.
Harris, R. S.
1968 Art and science of dental caries research. New York:
Academic Press.
Haydon, R. C.
1993 Survey of genetic variation among the Chiribaya of
the Osmore Drainage basin, southern Per. Ph.D.
dissertation, Department of Anthropology, Univer-
sity of Chicago.
Hoshower, L. M., J. E. Buikstra, P. S. Goldstein, and A. D.
Webster
1995 Articial cranial deformation at the Omo M10 Site:
A Tiwanaku complex from the Moquegua Valley,
Peru. Latin American Antiquity 6 (2): 145164.
Keyes, C. F.
1976 Towards a new formulation of the concept of ethnic
group. Ethnicity 3:202213.
Kolata, A.
1993 The Tiwanaku: Portrait of an Andean civilization.
Cambridge: Blackwell.
Lozada, M. C., D. Blom, and J. E. Buikstra
1996 Evaluating verticality through cranial deformation
patterns in the south Andes. Paper presented at
61st Annual Meeting of the Society for American
Archaeology, New Orleans.
Lozada, M. C., R. Haydon, and J. E. Buikstra
1993 Approaches to the study of genetic composition in
Late Intermediate period Chiribaya settlements,
southern Peru. Paper presented at 59th Annual
Meeting of the Society for American Archaeology,
St. Louis, Missouri.
Lukacs, J. R., and B. E. Hemphill
1991 The dental anthropology of prehistoric Baluchistan: A
morphometric approach to the peopling of south Asia.
In Advances in dental anthropology, edited by M. A. Kelly
and C. S. Larsen, 77120. New York: Wiley-Liss.
Menaker, L.
1980 Biological basis of dental caries, an oral biology textbook.
New York: Harper and Row.
Moore, W. J., and M. E. Corbett
1971 The distribution of dental caries in ancient British
populations. I. Anglo-Saxon period. Caries Res. 5:
151168.
Muoz, I.
1983 El poblamiento aldeano en el Valle de Azapa y su vin-
culacin con Tiwanaku (Arica-Chile). Documentos de
Trabajo 3: 4393. Arica: Instituto de Antropologa y
Arqueologa, Universidad de Tarapac.
Muoz, I., and G. Focacci
1985 San Lorenzo: Testimonio de una comunidad de
agricultores y pescadores postiwanaku en el Valle
de Azapa (Arica-Chile). Chungar 15:730.
Newburn, E.
1978 Cariology. Baltimore: Williams and Wilkins.
1982 Sugar and dental caries: A review of human studies.
Science 217:418423.
Oakland, A. R.
1992 Textiles and ethnicity: Tiwanaku in San Pedro de
Atacama, north Chile. Latin American Antiquity 3
(4): 316340.
Owen, B.
1993 A model of multiethnicity: State collapse, com-
petition, and social complexity from Tiwanaku to
Chiribaya in the Osmore Valley, Per. Ph.D. dis-
sertation, Department of Anthropology, University
of California, Los Angeles.
Powell, M. L.
1985 The analysis of dental wear and caries for dietary
reconstruction. In The analysis of prehistoric diets,
edited by R. I. Gilbert and J. H. Mielke, 307338.
New York: Academic Press.
Sackett, J.
1982 Approaches to style in lithic archaeology. Journal of
Anthropological Archaeology 1:59112.
Schiappacasse, V., A. Roman, I. Muoz, A. Deza, and G.
Focacci
1991 Cronologa por termoluminiscencia de la ceramica
del extremo norte de Chile: Primera parte. In Actas
de XI Congreso Nacional de Arqueologa Chilena, edited
by S. C. d. Arqueologa, 4359. Santiago: Museo
Nacional de Historia Natural.
Sjvold, T.
1973 The occurrence of minor non-metrical variants in
the skeleton and their quantitative treatment for
population comparisons. Homo 24:204233.
Soto-Heim, P.
1987 Evolucin de deformaciones intencionales, peina-
dos, tocados, y prcticas funerarias en la prehistoria
de Arica. Chungar 19:129214.
S U T T E R 205
Southall, A.
1976 Nuer and Dinka are real people: Ecology, ethnicity,
and logical possibility. Man 11 (4): 463491.
Stanish, C.
1989a An archaeological evaluation of an ethnohistorical
model in Moquegua. In Ecology, settlement, and his-
tory in the Osmore Drainage, Peru, edited by D. S.
Rice, C. Stanish, and P. R. Scarr, 303320. BAR
International Series 545. Oxford: British Archaeo-
logical Reports.
1989b Household archaeology: Testing models of zonal
complementarity in the south-central Andes.
American Anthropologist 91 (1): 724.
Stuiver, M., and P. J. Reimer
1993a CALIB users guide. Seattle: University of Washing-
ton Quaternary Research Center.
1993b Extended 14C database and revised CALIB radiocar-
bon calibration program. Radiocarbon 35:215230.
Sutter, R. C.
1994 Grupos bioculturales en los valles y costa de
Moquegua y Azapa: Verticalidad u horizontali-
dad? Paper presented at XII Congreso Nacional de
Arqueologa Chilena, Universidad de Antofogasta.
1995 Dental pathologies among inmates of the Monroe
County poorhouse. In Bodies of evidence: Reconstruct-
ing history through skeletal analysis, edited by A. L.
Grauer, 185196. New York: Wiley-Liss.
1996 A bioarchaeological perspective on verticality in
the middle and lower Moquegua Valley, Per, dur-
ing the Late Intermediate period. Paper presented
at Sixty-First Annual Meeting of the Society for
American Archaeology, New Orleans.
1997 Dental variation and biocultural afnities among
prehistoric populations from the coastal valleys of
Moquegua, Peru, and Azapa, Chile. Ph.D. disserta-
tion, Department of Anthropology, University of
Missouri.
2000 Prehistoric genetic and culture change: A bioar-
chaeological search for pre-Inka altiplano colonies
in the coastal valleys of Moquegua, Peru, and Azapa,
Chile. Latin American Antiquity 11:4370.
Sutter, R. C., and L. M. Mertz
2004 Non-metric cranial trait variation and prehistoric
biocultural change in the Azapa Valley, Chile. Amer-
ican Journal of Physical Anthropology 123:130145.
Turner, C. G.
1979 Dental anthropological indications of agriculture
among the Jomon people of central Japan. American
Journal of Physical Anthropology 51:619636.
Turner, C. G., II, C. R. Nichol, and G. R. Scott
1991 Scoring procedures for key morphological traits of
the permanent dentition: The Arizona State Uni-
versity dental anthropology system. In Advances in
dental anthropology, edited by M. A. Kelly and C. S.
Larsen, 1332. New York: Wiley-Liss.
206
INTRODUCTION
One of the most debated topics in archaeologi-
cal theory has been the term ethnicity. Since the
publication of Barths Ethnic Groups and Boundaries
in 1969, the concept of ethnicity has gone through
many transformations (Aldenderfer and Stanish 1993;
Conkey and Hastdorf 1990; Oakland Rodman 1992;
Shennan 1989; Wiessner 1983, among others). In a
way, these changes reect the difculty of creating a
universal denition of ethnicity. From the standpoint
of the archaeologist it may be more meaningful and
accurate to derive ethnicity from the specic context
in which it is being applied. Aldenderfer and Stanish
state: rather than focusing upon a potentially sterile
denition [of ethnicity], it is more valuable to look for
critical features of ethnic groups, and then to make
some effort to determinate to what extent ethnicity
can be read and interpreted from the archaeological
record (Aldenderfer and Stanish 1993:7).
In the case of prehispanic Peru, we are fortunate to
have unprecedented tools to explore the past, including
topics such as ethnicity. The archaeological record is
extremely rich, in part because the climate throughout
much of Peru has helped to preserve many details of
human activity not normally accessible in other parts
of the world. Of even greater importance, there is an
abundance of ethnohistorical accounts written during
and after the conquest that shed light on the unique
mosaic of social and cultural groups in the Andes,
including ethnic groups.
Beginning with one of her rst books, Curacas y Suce-
siones, Costa Norte in 1961, the Peruvian ethnohistorian
Mara Rostworowski de Diez Canseco has studied in
detail the social, economic, and cultural dynamics of
late prehispanic coastal populations in northern and
central Peru. After a detailed analysis of judicial and
administrative documents, she proposes that coastal
populations espoused a horizontal system of eco-
nomic organization, differing in many respects from
the vertical systems present in highland societies (Ros-
tworowski de Diez Canseco 1970). Rostworowski states
that, in contrast to the highland polities, prehispanic
CHAP T E R 1 2
Pescadores and Labradores among
the Seoro of Chiribaya in
Southern Peru
M. C. LOZADA
AND
J ANE E. BUI KS TRA
L O Z A D A A N D B U I K S T R A 207
coastal cultures did not expand into the altiplano to
obtain complementary resources from other eco-
logical niches, and that they developed independent
of highland influence. Furthermore, these coastal
polities, or seoros, were autonomous political and
economic groups composed of smaller sociopolitical
units or ethnic groups, under the control of their own
lords. Detailed descriptions in written accounts also
indicate that these ethnic groups, or parcialidades, were
economically premised on communities of specialists
(Rostworowski 1970, 1977, 1989, 1993).
We will focus on two fundamental ethnic groups
that, according to Rostworowski, served as the basis
of late prehispanic coastal societies: the shermen, or
pescadores, and farmers, or labradores. The existence of
communities of pescadores and labradores was initially
documented among the inhabitants of the Chincha
Valley in AVISO,
1
one of the rst colonial documents
that explicitly mentions the presence of occupa-
tional specialization within the seoro of Chincha
(Rostworowski 1970).
According to the census described in AVISO, the
pescadores were responsible for marine-based subsis-
tence activities (Rostworowski 1970):
There were communities along the coast consisting of
ten thousand pescadores that everyday went to the sea
each with their own boats and shing nets, and they
left and entered designated ports.
2
(AVISO, Rostworowski 1970:170)
In addition, the pescadores conducted other ancillary
activities such as drying and salting of sh for preserva-
tion (Rostworowski 1981). The existence of communities
of pescadores has been recorded in many other coastal
valleys throughout northern and central Peru (Hart
1983; Marcus 1987; Netherly 1977; Ramrez 1982;
Rostworowski 1981). Written sources indicate that the
pescadores were physically separated from the rest of
the seoro (Rostworwoski 1975a:315) and practiced
endogamy, such that there was essentially no intermar-
riage of pescadores with labradores. This practice was
severely tested shortly after the Spanish Conquest when
disease depopulated large areas of Peru. Women in sh-
ing communities would wait for shermen to come from
other areas instead of marrying members of nonshing
communities (Rostworowski 1970). This physical separa-
tion of communities continued well into the eighteenth
century in the port of Callao along Perus central coast,
testifying to its lasting influence among indigenous
coastal inhabitants (Flores Galindo 1981).
The separation of pescadores from other groups
within coastal societies was so fundamental that they even
possessed their own dialect, known as la lengua pescadora,
or shermens tongue, which served to distinguish them
from the labradores (Netherly 1977; Rabinowitz 1983;
Rostworowski 1981). Social distinctions also extended to
religious worship, as records show that pescadores prayed
to their own huacas
3
(sacred places) and gods, usually dur-
ing times that were determined by maritime activities.
They revered the moon in particular since it was known
to inuence the tides.
Labradores, on the other hand, were responsible for
cultivating the seoros lands. Furthermore, written
accounts indicate the labradores had the responsibility of
defending the lands situated at the head of the coastal val-
leys and guarding the water used for the irrigation of their
cultivated elds on the coast (Rostworowski 1990, 1995).
In coastal groups, the labradores were not only more
numerous than the pescadores, but generally possessed
greater inuence within the seoro as a whole. Rost-
worowski (1981) states that the pescadores depended on
the labradores, and, ultimately, the lord of the pescadores
was subject to that of the labradores. The same relation-
ship was observed by Hart among the inhabitants of the
Moche Valley in northern Peru: In general, specialist
groups seem to have been subordinate to a cacique princi-
pal whose close afliation was with agriculturalists (Hart
1983:253254). According to the historical accounts, each
parcialidad had its own elites, and these lords ultimately
were under the authority of a single paramount lord who
was afliated with the labradores.
These documents clearly suggest that parcialidades
of specialists were the building blocks of coastal societies
and represented discrete ethnic groups (Netherly 1977).
Archaeology represents an invaluable tool to evaluate the
social, political, and economic implications of this ethno-
historic model. The use of biological data derived from
archaeological materials, however, has not gured as
prominently in attempts to apply these models to specic
contexts. Human remains, in particular, possess a series
of physical and chemical attributes that yield unique data
on the lifestyle and biological identity of both individu-
als and populations as a whole (Buikstra and Cook 1980;
Cohen and Armelagos 1984; Iscan and Kennedy 1989;
Larsen 1997). Such biological data derived from human
remains permit one to analyze and comment on a broad
range of questions regarding ancient human behavior.
Migrational patterns, the local impact of colonization,
and the biological interaction between populations
can be detected through studies of genetic biodistance
208 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
based on cranial metric and non-metric traits and ancient
DNA. The study of intentional cranial deformation,
which served as a lifelong symbol of group membership,
is also useful to identify important social and ethnic
groups, as well as to study factors that determined their
residence behaviors and marriage practices (Hoshower
et al. 1995). Finally the reconstruction of ancient diets
based on an analysis of trace elements and stable isotopes
from human bone provides invaluable data concerning
subsistence patterns and resource utilization (Gilbert and
Mielke 1985; Sanford 1993).
This study addresses the implications of Rost-
worowskis coastal model as it applies to the Chiribaya
population of southern Peru (gure 1). The Chiribaya,
initially dened on the basis of ceramics, ourished
between AD 772 and 1350, centralizing their power in
the coastal region of the Osmore Valley, which is known
today as Ilo (Buikstra et al. 1997). This region repre-
sents an ideal forum in which to evaluate this model
for several reasons. First, the archaeological record of
the coast of Ilo demonstrates no evidence of highland
colonization, even during periods when the Tiwanaku
state and the Inka Empire controlled the Osmore mid-
valley (Bawden 1989; Owen 1993; Stanish 1992). In this
respect the populations that inhabited the coast of Ilo,
such as Chiribaya, are more likely to represent purely
coastal societies, similar to those described by Rost-
worowski along the northern coasts of Peru (Bawden
1989; Buikstra 1995). Second, the Ilo region has a
long history of occupation, with evidence of human
habitations as early as 10,575 +/-105 BP (Sandweiss et
al. 1989). This research integrates both biological and
cultural data from the coastal polity of Chiribaya and
addresses the following questions:
Figure 1. South-central Andes.
L O Z A D A A N D B U I K S T R A 209
1. Were the Chiribaya biologically independent of
highland traditions?
2. Was Chiribaya made up of endogamous commu-
nities of economic specialists?
3. Did Chiribaya represent a seoro, a confederacy of
economically specialized parcialidades, integrated
under the authority of a single paramount lord?
In order to evaluate these hypotheses, two types of bio-
archaeological analysis will be used: biological distances
through non-metric cranial traits and intentional cranial
deformation styles. The estimation of genetic relation-
ships in skeletal populations has been used frequently to
evaluate ancient migrational and/or residential patterns.
In particular, the study of epigenetic traits has been the
most commonly used method to measure genetic differ-
ences in the Andes (Blom et al. 1998; Chakraborty et al.
1976; Cocilovo and Rothhammer 1984, 1990; Cocilovo
et al. 1982; Guilln 1992; Lozada, Haydon, and Buikstra
1993; Rivera and Rothhammer 1986; Rothhammer et al.
1981; Rothhammer et al. 1982, 1984, 1989; Rothhammer
et al. 1983; Rothhammer et al. 1986; Rothhammer and
Silva 1989, 1990; Soto and Rothhammer 1975; Soto et
al. 1975; Sutter 1997, 2000).
As the name implies, epigenetic traits are not direct
measurements of genotype, but instead, phenotypic
traits with high heritability (Hauser and De Stefano
1989). These traits largely represent minor variations in
bony development (Berry and Berry 1963). To the extent
that such traits are determined by genetic as opposed to
environmental factors, they represent an accurate way to
estimate genetic relatedness (Buikstra 1976).
Bioanthropologists have selected those traits with a
high degree of heritability, representing the best reec-
tion of genotype (Cheverud and Buikstra 1981a, 1981b;
Ishida and Dodo 1997; Hauser and De Stefano 1989;
Johnson, Gorlin, and Anderson 1965; Pietrusewsky and
Douglas 1993; Saunders and Popovich 1978; Susuki and
Sakai 1960; Torgersen 1951). These traits have been
classied into four groups based on their etiology: 1)
hyperostotic traits: ossication of soft tissue structures
such as ligaments/tendons; 2) hypo-ostotic traits:
areas of incomplete ossication; 3) foramina (osseous
tunnels)/grooves: variations in the osseous channel for
neurovascular bundles; and 4) supernumerary vault
sutures and ossicles. Approximately 200 such variations
have been identied in the skull, and another 200 in the
postcranial skeleton (Saunders 1989).
Samples included in this study come from San
Gernimo, El Yaral, and Chiribaya Alta (figure 2).
Radiocarbon dates have been obtained from each of
these sites (Buikstra et al. 1997) and indicate that there
was substantial chronological overlap between all three
Figure 2. Map of the Osmore Drainage and site location.
210 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Figure 3. Map of the mortuary component of San Gernimo.
L O Z A D A A N D B U I K S T R A 211
sites, although dates from San Gernimo begin slightly
later than Yaral and Chiribaya Alta. San Gernimo is
the most extensive Chiribaya site on the Ilo coast,
located only a hundred meters from the sea (gure
3). The individuals recovered from this site have been
characterized as the pescadores of Chiribaya (Buikstra
1995). As such, they were buried with their tools of
production such as sh hooks, harpoons, weights, nets,
ropes, oats, and miniature boats (Jessup 1990a, 1990b).
Large deposits containing anchovy were also excavated,
as well as middens with large quantities of shells, algae,
and sea mammal bones, reflecting the importance
of marine resources in the subsistence base of San
Gernimo (Jessup 1990b). This dependence on marine
exploitation is further corroborated by analyses of stable
isotopes and trace elements from San Gernimo. On
the basis of these analyses, Tomczak (1995) states that
the inhabitants of San Gernimo subsisted on a diet
high in marine consumption.
In contrast, the site of El Yaral is located fty kilome-
ters away from the sea at 1,000 masl (meters above sea
level) and associated with irrigation canals and agricul-
tural terraces (Buikstra 1988; Jessup 1987; Rice 1993;
Stanish 1992) (gure 4). Virtually no artifacts associ-
ated with shing activities or marine consumption have
been detected in this community. Economic activities
inferred from domestic and mortuary excavations sug-
gest a heavy reliance on terrestrial resources, especially
agricultural ones, and attest to the agricultural special-
ization by individuals buried at El Yaral. Tomczaks
dietary studies conrm that the subsistence base of
the labradores at this site relied much more heavily on
terrestrial agricultural products than did that of coastal
Chiribaya communities.
Finally, the site of Chiribaya Alta, located seven
kilometers from the coast on a promontory known as
the Pampa del Descanso, represents the most extensive
and complex Chiribaya site in southern Peru (gure 5).
The site covers approximately thirty-six hectares and is
formed by a large mortuary component that includes at
least nine bounded cemeteries and limited residential
areas. The complex organization of Chiribaya Alta, as
well as its role within the Chiribaya polity, is incom-
pletely understood. Considerable overlap occurred
between the usage of cemeteries at Chiribaya Alta,
indicating that many cemeteries were in use at the
same time; however, there are noticeable differences
in mortuary assemblages between cemeteries, raising
the possibility that they were used by different ethnic
groups (Lozada, Blom, and Buikstra 1997). Given the
lack of a large residential component, and the relatively
opulent burial offerings associated with tombs from
Chiribaya Alta, Lozada and colleagues (1997) suggested
that Chiribaya Alta served as a center for ceremonial
activity and political power, in much the same way as
Pachacamac (Shimada 1991) in the central coast and
Pacatnam in the north coast (Donnan 1997).
A high degree of variability in mortuary patterns,
ceramic styles, as well as dietary staples are present at
this site, reecting its role as the crossroads of Chirib-
aya. Furthermore, the different cemeteries of Chiri-
baya Alta demonstrate close afliations with specic
satellite communities. Individuals buried in cemetery
4, for example, exhibit close ties with the pescadores
of San Gernimo. In contrast, cemeteries 1, 2, 7, and
9 are more similar to inland Chiribaya sites associated
with agricultural activities, both in material culture and
dietary patterns. Additional data were collected from
three other sites for comparison: 1) the Chen Chen site
of the Osmore mid-valley, associated with the Tiwanaku
V phase, currently hypothesized as the biological pre-
cursor of Chiribaya; 2) the contemporaneous Tumilaca
site of El Algodonal; and 3) the Formative site of Roca
Verde, located on the coast of Ilo, dated to approxi-
mately 500 BC to AD 200.
CHIRIBAYA BIODISTANCE ANALYSIS
Non-metric traits were scored and analyzed using stan-
dard methodologies in physical anthropology (Buikstra
and Ubelaker 1994) and used by Blom in her analysis
of Chen Chen (Blom et al. 1998). In this study, the bio-
distance between samples was determined statistically
by using the mean difference of the trait frequencies:
where D is biodistance, f1i and f2i are frequencies
on the ith trait for populations 1 and 2, and n is the
number of traits. To calculate the probability that each
distance was obtained by chance in samples drawn from
a homogeneous population, a bootstrap method was
used (Manly 1991). In this method, the data set was
subjected to a series of 5001,000 randomizations in
which each variable was independently sorted according
to random numbers. This had the effect of randomizing
the assignment of values to individuals while maintain-
ing the probability distribution of trait values for the
whole sample, and thus simulates the null hypothesis
that the samples were drawn from the same population.
Dendograms were then constructed using parsimony
analysis. In this method, we compared the parsimony
of the reasonable alternatives that were obtained by
212 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Figure 4. Site of El Yaral.
L O Z A D A A N D B U I K S T R A 213
pairing samples with their closest neighbor. The set
of distances was thereby successively decomposed into
the closest pairs until all groups had been assigned to a
clade. In the case of several reasonable alternatives, the
route of least additive distance was selected.
The results of the biodistance analysis between
Chen Chen, different Chiribaya cemeteries, El Algo-
donal, and Roca Verde are recorded in table 1. The
matrix of biological distance reveals a clear pattern
among the samples analyzed. Specically, the larg-
est distances occurred between the Chen Chen and
Chiribaya samples, and between Chen Chen and El
Algodonal. There were no statistically significant
differences between individual Chiribaya cemeteries
linked with pescadores such as San Gernimo and the
labradores in Yaral. In fact they appear to be biologically
homogeneous and not genetically isolated as has been
postulated by Rostworowski ethnohistorically. Interest-
ingly, the values between the Chiribaya and Tumilaca
exhibit no distinguishable differences.
In table 1 one can observe the distance matrix after
all Chiribaya samples were pooled and compared to El
Algodonal, Chen Chen, and Roca Verde. Once again, a
statistically signicant biological distance occurs between
Chen Chen and Chiribaya, as well as between Chen
Chen and El Algodonal. Comparisons between Chen
Chen and Roca Verde yielded a p value of 0.063, which
is on the cusp of the 0.05 level of probability. Given the
relatively small sample sizes in Roca Verde, this probabil-
ity may be a reection of sampling, and therefore should
be tested upon a larger sample. Overall, the data suggest
a relatively distant biological relationship between the
coastal populations that inhabited the coast of Ilo and the
Tiwanaku colonists at Chen Chen. In contrast, the Roca
Verde and the Chiribaya samples are genetically indistin-
guishable, reecting biological continuity between these
Figure 5. Site of Chiribaya Alta.
214 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
two coastal samples. These results are presented in the
form of a dendogram in gure 6. Here, the biological
relationships among these various samples are further
illustrated. When each individual was grouped accord-
ing to ceramic association, a similar distance matrix was
observed (see table 1). Larger genetic distances were
found between Chiribaya, Ilo/Tumilaca, and Chen
Chen samples. However, there are no major differences
between Chiribaya, Tumilaca, and Roca Verde.
In summary, the Chen Chen skeletal sample is
biologically different from the Chiribaya, Ilo-Tumilaca/
Cabuza, and Roca Verde samples. In this sense, Chiri-
baya does not appear to be biologically continuous
with Tiwanaku colonies in the Osmore Drainage in
southern Peru. On the contrary, there is closer bio-
logical affinity between Chiribaya and the coastal
Formative site of Roca Verde. In addition, no internal
biological patterning was detected among the Chiribaya
samples, underscoring the lack of genetic isolation (i.e.
endogamy) between the pescadores and labradores. It
remains unclear whether epigenetic traits are sensitive
enough to detect endogamy within small populations;
however, the lack of measurable genetic differences
between Chiribaya cemeteries echoes similar results
found by Haydon (1993) on the basis of microsatellite
ancient DNA. This remarkable degree of phenotypic
and genotypic homogeneity across Chiribaya com-
munities indicates that complete genetic isolation was
unlikely; however it does not completely disprove the
practice of endogamy among the Chiribaya. Additional
ethnohistoric accounts may help to explain this pattern.
Although marriages between groups of specialists have
not been documented, there are instances in which
lords would marry the daughter of another lord, even
though they were members of different groups, as part
of marriage alliances (Hart 1983). This practice proba-
bly served important political functions in that it helped
to promote integration between parcialidades.
Another potential cause of gene ow, in spite of the
practice of endogamy, may be found in the ancient prac-
tice of tinkuys, an annual ritual battle between different
moieties within a community:
Annual intergroup duels between men, and occasion-
ally between women, are commonplace throughout the
Andes. Their generic Quechua name is tinkuy, meaning
an encounter between two like entities or a combina-
tion of two substances. They vary in their ferocity, but
in many of them deaths are not unusual. Some also
involve the capture and rape of women. The span of
the participating groups varies from case to case. They
may be two sections of a single village or town, or of
an ethnic group, or they may be coalitions of commu-
nities in a locality or region. . . . Women attended to
animate the men and to rouse them to feats of bravery,
but risked abduction, being regarded as potential brides
of their captors.
(Sallnow 1991:298299)
Chen Chen Algodonal Chiribaya
0.063*
0.111*
Roca Verde
0.149, p<0.001
* Statistically significant for =0.05
Figure 6. Dendogram of biological distances by sites.
L O Z A D A A N D B U I K S T R A 215
Y
a
r
Y
a
r
S
C
H
A
1
C
H
A
C
H
A
3
C
H
A
4
C
H
A
7
C
h
e
n
C
h
e
n
R
V
A
l
g
o
d
o
n
a
l
D
=
0
.
1
3
D
=
0
.
1
0
D
=
0
.
0
7
D
=
0
.
0
8
D
=
0
.
1
3
D
=
0
.
0
8
D
=
0
.
0
9
D
=
0
.
1
0
D
=
0
.
1
5
D
=
0
.
1
1
p
=
0
.
3
3
p
=
0
.
6
9
p
=
0
.
8
9
p
=
0
.
8
8
p
=
0
.
2
3
p
=
0
.
8
8
p
=
0
.
8
6
p
=
0
.
6
8
p
<
0
.
0
0
p
=
0
.
6
6
Y
a
r
a
l
D
=
0
.
1
0
D
=
0
.
0
9
D
=
0
.
1
1
D
=
0
.
1
2
D
=
0
.
1
1
D
=
0
.
0
8
D
=
0
.
1
2
D
=
0
.
2
0
D
=
0
.
1
5
p
=
0
.
8
6
p
=
0
.
7
8
p
=
0
.
6
7
p
=
0
.
5
1
p
=
0
.
5
2
p
=
0
.
9
6
p
=
0
.
5
6
p
<
0
.
0
0
p
=
0
.
3
4
Y
a
r
a
l
D
=
0
.
1
0
D
=
0
.
1
1
D
=
0
.
1
4
D
=
0
.
1
2
D
=
0
.
0
7
D
=
0
.
1
2
D
=
0
.
1
6
D
=
0
.
1
1
p
=
0
.
6
5
p
=
0
.
5
7
p
=
0
.
2
2
p
=
0
.
3
5
p
=
0
.
9
8
p
=
0
.
4
8
p
<
0
.
0
0
p
=
0
.
8
5
S
a
n
D
=
0
.
1
0
D
=
0
.
1
0
D
=
0
.
1
0
D
=
0
.
0
8
D
=
0
.
0
9
D
=
0
.
1
5
D
=
0
.
1
3
p
=
0
.
4
7
p
=
0
.
6
1
p
=
0
.
3
9
p
=
0
.
8
8
p
=
0
.
6
8
p
<
0
.
0
0
p
=
0
.
3
3
C
H
A
1
D
=
0
.
1
3
D
=
0
.
1
1
D
=
0
.
1
1
D
=
0
.
1
2
D
=
0
.
1
4
D
=
0
.
1
3
p
=
0
.
3
2
p
=
0
.
4
3
p
=
0
.
6
5
p
=
0
.
3
4
p
<
0
.
0
0
p
=
0
.
4
3
C
H
A
2
D
=
0
.
1
2
D
=
0
.
1
6
D
=
0
.
1
1
D
=
0
.
1
9
D
=
0
.
1
5
p
=
0
.
2
p
=
0
.
0
4
p
=
0
.
5
5
p
<
0
.
0
0
p
=
0
.
3
3
C
H
A
3
D
=
0
.
1
1
D
=
0
.
1
3
D
=
0
.
1
4
D
=
0
.
1
0
p
=
0
.
4
7
p
=
0
.
0
9
p
<
0
.
0
0
p
=
0
.
8
4
C
H
A
4
D
=
0
.
1
0
D
=
0
.
1
3
D
=
0
.
1
2
p
=
0
.
8
4
p
=
0
.
0
2
p
=
0
.
7
1
C
H
A
7
D
=
0
.
1
7
D
=
0
.
1
6
p
<
0
.
0
0
p
=
0
.
1
4
C
h
e
n
D
=
0
.
1
4
p
=
0
.
0
6
T
a
b
l
e
1
.
M
a
t
r
i
x
o
f
b
i
o
l
o
g
i
c
a
l
d
i
s
t
a
n
c
e
b
e
t
w
e
e
n
i
n
d
i
v
i
d
u
a
l
c
e
m
e
t
e
r
i
e
s
.
216 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
In this respect, the ritual conict between social moi-
eties often resulted in abduction and rape of women,
possibly from other communities. Therefore, even
though endogamy may have been an ideal marriage
pattern, there existed within Andean customs various
means of circumventing it. If such rituals were also
practiced among the Chiribaya, this custom may have
been responsible for signicant amounts of gene ow
that would have largely eliminated genetic differences
between communities.
CRANIAL DEFORMATION PATTERNS WITHIN
CHIRIBAYA
The evaluation of cranial deformation among the Chiri-
baya dovetails nicely with the genetic data to provide
much needed insight into the internal composition and
structure of Chiribaya. Because cranial deformation is
performed when a person is born, it served as a life-long
symbol of group membership. It often required several
years of dedicated effort to produce a certain head
form, and therefore, represented an important invest-
ment of labor from the community. Chroniclers such
as Bartolom de las Casas (1892 [ca. 1560]), Bernab
Cobo (1956 [1653]), and Garcilaso de la Vega (1991
[1609]) indicate that cranial deformation was used prin-
cipally to create visual cues regarding an individuals
ethnic or status afliation. This practice accompanied
other means of differentiating oneself from other
groups such as clothing, bodily adornments, and other
material accessories; however, an important distinction
must be made between the use of cranial deformation
and material culture as symbols of an individuals social
personae. Cranial deformation was imposed on an
individual at birth, and could not be altered despite
subsequent changes in an individuals life. As such,
cranial deformation represents a symbol of ascribed
identity. In contrast, ceramics, textiles, and other items
that incorporate stylistic symbols of group differences
were subject to personal preferences and day-to-day
variation, and therefore more likely reected achieved
identity. The study of both achieved and ascribed com-
ponents of an individuals social personae is important
in analyses of social structure, and as such, cranial defor-
mation will help to complement contextual data derived
from other forms of material culture.
The classication and description of cranial defor-
mation styles among the Chiribaya was done using
methodologies developed by Hoshower et al. (1995)
and Blom, Yeatts, and Buikstra (1995) for the Omo and
Chen Chen collections, respectively. This consisted of
scoring each skull according to criteria related to their
shape and deformation techniques. Throughout this
investigation, forms designed by Blom and colleagues
(1995) were used. They include specic information
about cranial morphology, areas affected by the defor-
mation apparatus, and impressions left by the elements
used to achieve the deformation, such as pads, tablets,
and straps.
In this analysis, two broad deformation groups were
distinguished. The rst type was the tabula obliqua,
described by Dembo and Imbellioni (1938), character-
ized by fronto-occipital compression (gure 7). On the
basis of this study, this category was further divided
into three types: 1) fronto-occipital at, 2) fronto-
occipital round, and 3) fronto-occipital slight (gure
8). The second group consists of the circumferential
(Hrdlicka 1912), or annular obliqua, type (Dembo and
Imbellioni 1938), characterized by an elongated and
tubular vault (gure 9). In both groups, the orienta-
tion of the vault, as reected by the angulation of the
frontal bone, is oblique and not vertical or erect, as has
been documented in other pre-Columbian populations
(Allison et al. 1981).
As can be seen in figure 10, 40% of the sample
was not intentionally deformed. The remaining 60%
of skulls were deformed, and were divided into 20%
annular, 23% fronto-occipital/at, 9% fronto-occipital/
round, and 8% fronto-occipital/slight, respectively.
Based on these results, the fronto-occipital types of
cranial deformation were twice as common as the
annular forms. In order to interpret more specically
the patterning of these deformation types, the cor-
relation between these types and other biological and
cultural variables such as biological sex, burial location,
and ceramic association was conducted.
No associations can be seen between cranial defor-
mation styles and skeletal sex. Therefore, cranial
deformation was not used to display or symbolize gen-
der identity. Furthermore, cranial deformation does
not appear to have been used to symbolize status, even
though it was more common among elite burials. By far
the most important determinant of cranial deformation
styles was economic specialization, as reected in their
distribution by cemetery. The fronto-occipital styles
of deformation demonstrate clear association with the
inland site of El Yaral, as well as with those cemeteries
at Chiribaya Alta, which were closely afliated with the
labradores (gure 11). Conversely, the annular style was
closely linked to the pescadores at San Gernimo and
L O Z A D A A N D B U I K S T R A 217
Figure 7. Fronto-occipital, or tabula obliqua, type. Figure 9. Circumferential, or annular, type.
Figure 8. Cranial deformation types.
(Lozada, Blom, and Buikstra 1997)
218 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
0
10
20
30
40
50
Fronto-
Occipital Slight
Fronto-
Occipital
Round
Normal Fronto-
Occipital Flat
Circumferential/
Annular
Figure 10: Frequency of Cranial Deformation Types
Figure 10. Frequency of cranial deformation types.
0
20
40
60
80
A
l
g
o
d
o
n
a
l
Y
a
r
a
l
1
Y
a
r
a
l
2
C
H
A
1
C
H
A
2
C
H
A
3
C
H
A
4
C
H
A
5
C
H
A
6
C
H
A
7
S
a
n
G
e
r
n
i
m
o
Figure 11: Frequency of Fronto-Occipital Flat Deformation by Cemetery
Figure 11. Frequency of fronto-occipital at deformation within cemeteries.
0
20
40
60
80
A
l
g
o
d
o
n
a
l
Y
a
r
a
l
1
Y
a
r
a
l
2
C
H
A
1
C
H
A
2
C
H
A
3
C
H
A
4
C
H
A
5
C
H
A
6
C
H
A
7
S
a
n
G
e
r
n
i
m
o
Figure 12: Frequency of Circumferential/Annular Deformation by Cemetery
Figure 12. Frequency of annular deformation by cemetery.
L O Z A D A A N D B U I K S T R A 219
Chiribaya Alta 4 (CHA-4) (gure 12). The temporal
overlap between sites, as indicated by previous radio-
carbon dating, indicates that differences in the practice
of cranial deformation were not due to changes over
time. Using cranial deformation styles, therefore, one
can identify ethnically distinct groups within Chiribaya,
which correspond to groups of economic specialists
such as pescadores and labradores. Furthermore, cra-
nial deformation styles can be shown to correlate well
to other cultural practices such as ceramic decoration
(gure 13) and tomb location (gure 14). In this respect,
the Chiribaya had a clear sense of their own identity,
and this was manifested in various cultural mediums
such as cranial deformation.
Ethnohistorically, pescadores symbolized their
separation from other communities in a variety of ways
such as language and religious ideology (Netherly 1977;
Rabinowitz 1983; Rostworowski 1975a, 1981), and we
would propose that cranial deformation was yet another
means by which they visibly distinguished themselves
from other groups. Cranial deformation was, therefore,
part of a cluster of distinct cultural practices includ-
ing subsistence base, ceramic decoration, and burial
program that served to reinforce and symbolize clear
social and cultural differences. In a similar fashion, the
fronto-occipital/at style of cranial deformation sym-
bolized the labradores of Chiribaya. Fronto-occipital
at skulls are noticeably absent from the site of San
Gernimo, and become more common as one moves
up the valley, the reverse of the pattern found with the
annular style. They are most prevalent within cemetery
7 at Chiribaya Alta and cemetery 2 at El Yaral. Burials
associated with this type of cranial deformation were
also highly associated with Algarrobal ceramics and
burial offerings of camelid remains. These features
contrast with those identified with the pescadores
summarized above, and appear to indicate that a very
different set of symbols was utilized by the labradores to
identify themselves. The remaining deformation styles
such as fronto-occipital/slight represent relatively subtle
modications of skull shape and are found principally
in inland communities, whose subsistence base was
increasingly from nonmarine resources. In this respect,
the technique of fronto-occipital compression is closely
linked with labradores.
This pattern is consistent with the descriptions of
Rostworowski for northern and central coastal popu-
lations. She described groups of loosely integrated
communities that maintained very separate identities,
and were relatively equivalent units within the larger
political landscape. These parcialidades served as the
building blocks of coastal seoros, and this appears to
have been equally true among the Chiribaya.
It is also possible on the basis of archaeological data
and cranial deformation styles to postulate the existence
of a paramount lord among the Chiribaya. In particular,
burial 419 at Chiribaya Alta is highly distinct from other
tombs, reecting the unique position of its occupants.
For example, it is the only instance in which three indi-
viduals were buried together in the same tomb. In this
case, two women and a single male. The investment of
labor in this burial was truly monumental by Chiribaya
standards. Of even greater interest, however, is the fact
that the male has a cranial deformation style similar to
those of the labradores, while the two females had the
annular type associated with the pescadores. Given that
the paramount lord was entrusted with the duty of inte-
grating the disparate communities within the seoro,
this highly visible association between groups of eco-
nomic specialists must have served as a powerful symbol
of intergroup solidarity. Furthermore, the placement of
burial 419 close to the center of Chiribaya Alta, which
can be viewed as a microcosm of the seoro as a whole,
helps to reinforce the central role of this lord within the
sociopolitical consciousness of Chiribaya.
In light of these data, the perception of coastal
environments as less conducive to the development
of complex societies, therefore, must be challenged.
Owen (1993) and Stanish (1992) postulated that
Chiribaya developed from the remnants of Tiwanaku:
however, we would argue that they originated from
prior coastal populations at the height of Tiwanakus
power in the Osmore mid-valley. In this respect, the
Chiribaya were not merely the cultural beneciaries
of highland societies, but rather independent foci of
cultural development that remained buffered from
mid-valley colonies and eventually outlived Tiwanakus
presence. Furthermore, their social organization along
ethnic lines closely follows the ethnohistorical model
proposed by Rostworowski for coastal communities in
northern and central Peru. Thus denitions of ethnicity
and the identication of ethnic groups in archaeological
contexts are greatly strengthened by the use of ethno-
historical analysis in conjunction with human biological
and archaeological data.
220 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
0
2
0
4
0
6
0
8
0
1
0
0
T
u
m
i
l
a
c
a
Y
a
r
a
l
A
l
g
a
r
r
o
b
a
l
S
a
n
G
e
r
n
i
m
o
F
r
o
n
t
o
-
O
c
c
i
p
i
t
a
l
S
l
i
g
h
t
F
r
o
n
t
o
-
O
c
c
i
p
i
t
a
l
R
o
u
n
d
N
o
r
m
a
l
F
r
o
n
t
o
-
O
c
c
i
p
i
t
a
l
F
l
a
t
C
i
r
c
u
m
f
e
r
e
n
t
i
a
l
/
A
n
n
u
l
a
r
F
i
g
u
r
e
1
3
:
F
r
e
q
u
e
n
c
y
o
f
D
e
f
o
r
m
a
t
i
o
n
T
y
p
e
s
b
y
C
e
r
a
m
i
c
S
t
y
l
e
s
F
i
g
u
r
e
1
3
.
F
r
e
q
u
e
n
c
y
o
f
d
e
f
o
r
m
a
t
i
o
n
t
y
p
e
b
y
c
e
r
a
m
i
c
s
t
y
l
e
.
0
2
0
4
0
6
0
8
0
1
0
0
Algodonal
Yaral1
Yaral2
CHA1
CHA2
CHA3
CHA4
CHA5
CHA6
CHA7
San Gernimo
F
r
o
n
t
o
-
O
c
c
i
p
i
t
a
l
S
l
i
g
h
t
F
r
o
n
t
o
-
O
c
c
i
p
i
t
a
l
R
o
u
n
d
N
o
r
m
a
l
F
r
o
n
t
o
-
O
c
c
i
p
i
t
a
l
F
l
a
t
C
i
r
c
u
m
f
e
r
e
n
t
i
a
l
/
A
n
n
u
l
a
r
F
i
g
u
r
e
1
4
:
F
r
e
q
u
e
n
c
y
o
f
D
e
f
o
r
m
a
t
i
o
n
T
y
p
e
s
b
y
C
e
m
e
t
e
r
y
F
i
g
u
r
e
1
4
.
F
r
e
q
u
e
n
c
y
o
f
d
e
f
o
r
m
a
t
i
o
n
t
y
p
e
s
b
y
c
e
m
e
t
e
r
y
.
L O Z A D A A N D B U I K S T R A 221
NOTES
1. The author and date of this colonial document are
unknown. However, Rostworowski (1970) suggests that
AVISO was written after 1570 by Fray Pablo de Castro.
2. Ava poblados por la costa de la mar diez mil pesca-
dores, que cada da o los ms de la semana entravan en la
mar, cada uno con su balsa y redes y salan y entraban en sus
puertos sealados.
3. Huaca is a word for a place, person, or object consid-
ered to be sacred. In coastal areas, islands were important
huacas. According to yunga (people who inhabit either the
coast or mid-valley regions) origin myths, ancient lords were
transformed into islands by the mythical hero Vichama (Ros-
tworwoski 1997:35).
REFERENCES
Aldenderfer, M. S., and C. Stanish
1993 Domestic architecture, household archaeology,
and the past in the south-central Andes. In Domes-
tic architecture, ethnicity, and complementarity in the
south-central Andes, edited by M. S. Aldenderfer,
112. Iowa City: University of Iowa Press.
Allison, M., E. Gerszten, J. Munizaga, C. Santoro, and G.
Focacci
1981 La prctica de la deformacin craneana entre los pueb-
los Andinos precolombinos. Chungar 7:238260.
Anton, S.
1989 Intentional cranial vault deformation and induced
changes of the cranial base and face. American Jour-
nal of Physical Anthropology 79:253267.
Barth, F.
1969 Ethnic groups and boundaries: The social organiza-
tion of culture difference. Boston: Little, Brown and
Company.
Bawden, G.
1989 Settlement survey and ecological dynamics on the
Peruvian south coast. Andean Past 2:3967.
Berry, R. J., and A. Berry
1963 Epigenetic polymorphism in wild populations of
Mus musculus. Genetic Research 4:193220.
Blom, D. E., B. Hallgrmsson, L. Keng, M. C. Lozada, and
J. E. Buikstra
1998 Tiwanaku state colonization: Bioarchaeological
implications for migration in the Moquegua Valley,
Peru. World Archaeology 30 (2): 238261.
Blom, D. E., M. C. Lozada, B. Hallgrmsson, L. Keng, and
J. E. Buikstra
1997 Biology and culture in the prehispanic south-cen-
tral Andes. Paper presented at the 62nd Annual
Meeting of the Society for American Archaeology,
Nashville.
Blom, D. E., D. J. Yeatts, and J. E. Buikstra
1995 A bioanthropological approach to Tiwanaku-
Moquegua Valley interaction. Paper presented at
the 58th Annual Meeting of the Society for Ameri-
can Archaeology, Minneapolis.
Buikstra, J. E
1976 Hopewell in the lower Illinois Valley: A regional approach
to the biological variability and prehistoric behavior.
Northwestern University Archaeological Program
Scientic Papers 2. Evanston, IL.
1988 Chiribaya: An integrated approach to the biologi-
cal and cultural development of an Andean society.
Proposal for research support to the Archaeology
Program, National Science Foundation.
1995 Tombs for the living. . .or. . .for the dead: The
Osmore ancestors. In Tombs for the living: Andean
mortuary practices, edited by T. D. Dillehay, 229280.
Washington, D.C.: Dumbarton Oaks Research
Library and Collection.
Buikstra, J. E., and D. C. Cook
1980 Paleopathology: An American account. Annual
Review of Anthropology 9:433470.
Buikstra, J. E., M. C. Lozada, G. F. Rakita, and P. D. Tomczak
1997 Fechados radiocarbnicos de sitios Chiribaya en
la cuenca del Ro Osmore. Paper presented at
Seminario sobre Arqueologa de Ilo, Pontificia
Universidad Catlica del Per. Lima.
Buikstra, J. E., and D. Ubelaker, eds.
1994 Standards for data collection from human skeletal
remains. Arkansas Archaeological Survey Research
Series 44. Fayetteville.
Casas, B. de las
1892 [ca. 1560] De las antiguas gentes del Peru. Madrid: Tipo-
grafa de Manuel G. Hernandes.
Chakraborty R., R. Blanco, F. Rothhammer, and E. Llop
1976 Genetic variability in Chilean populations and its
association with geography, language, and culture.
Social Biology 23:7381.
Cheverud, J. M. and J. E. Buikstra
1981a Quantitative genetics of skeletal non-metric traits
in the rhesus macaques on Cayo Santiago. I. Single
trait heritabilities. American Journal of Physical
Anthropology 54:4349.
1981b Quantitative genetics of skeletal non-metric traits
in the rhesus macaques on Cayo Santiago. II. Phe-
notypic, genetic, and environmental correlations
between traits. American Journal of Physical Anthro-
pology 54:5158.
222 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Cobo, B.
1956 [1653] Obras del Padre Bernab Cobo. Biblioteca de
Autores Espaoles 9192. Madrid.
Cocilovo, J. A., and F. Rothhammer
1984 Relaciones y anidades biolgicas del grupo humano
de Camarones 14. In Descripcin y anlisis intepre-
tativo de un sitio arcaico temprano en la Quebrada de
Camarones, edited by V. Schiappacasse and H. Nie-
meyer, 52162. Santiago de Chile: Publicaciones
Ocasionales.
1990 Paleopopulation biology of the southern Andes:
Craniofacial chronological and geographical dif-
ferentiation. Homo 41 (1): 1631.
Cocilovo, J. A., F. Rothhammer, S. Quevedo, and E. Llop
1982 Microevolucin en poblaciones prehistricas del
rea Andina. III. La poblacin del Morro de Arica.
craneometra. Revista Universidad Nacional de Ro
Cuarto 2 (2): 91111.
Cohen, M. N., and G. Armelagos
1984 Paleopathology and the origins of agriculture. New York:
Academic Press.
Conkey, M. W., and C. Hastdorf, eds.
1990 The uses of style in archaeology. Cambridge: Cam-
bridge University Press.
Dembo, A., and J. Imbellioni
1938 Deformaciones intencionales del cuerpo humano.
Humanior: Biblioteca del Americanista Moderno,
Buenos Aires: Jos Anesi.
Donnan, C.
1997 Introduction. In The Pacatnamu papers, edited by C.
Donnan and G. Cook, 2:916. Los Angeles: Fowler
Museum of Cultural History, UCLA.
El-Najjar, M., and G. Dawson
1977 The effect of articial cranial deformation on the
incidence of wormian bones in the lambdoidal
suture. American Journal of Physical Anthropology 46:
155160.
Espinoza, S. W.
1975 El Valle de Jayanca en el reino de los Mochica siglos
XV y XVI. Boletn del Instituto de Francs de Estudios
Andinos 4 (34): 243274.
Feldman, R. A.
1989 The early ceramic periods of Moquegua. In Ecology,
settlement, and history in the Osmore Drainage, Peru,
edited by D. S. Rice, C. Stanish,and P. R. Scarr,
227235. BAR International Series 545 (ii). Oxford:
British Archaeological Reports.
Flores Galindo, A.
1981 La pesca y los pescadores en la costa central (s. XVIII).
Histrica 5 (2): 159167.
Garcilaso de la Vega
1991 [1609] Comentarios reales de los Incas. Vols. 12. Lima:
Fondo de Cultura Econmica.
Gilbert, R. I., and J. H. Mielke, eds.
1985 The analysis of prehistoric diets. Orlando: Academic
Press, Inc.
Guilln, S. E.
1992 The Chinchorro culture: Mummies and crania in the
reconstruction of preceramic coastal adaptation in
the south-central Andes. Ph.D. dissertation, Depart-
ment of Anthropology, University of Michigan.
Hart, E. A.
1983 Prehispanic political organization of the Peru-
vian north coast. Ph.D. dissertation, Department
of Anthropology, University of Michigan, Ann
Arbor.
Hauser, G., and G. F. De Stefano
1989 Epigenetic variants of the human skull. Stuttgart: Sch-
weizerbart.
Haydon, R.
1993 Survey of genetic variation among the Chiribaya of
the Osmore Drainage basin, southern Per. Ph.D.
dissertation, Department of Anthropology, Univer-
sity of Chicago.
Hoshower, L., J. E. Buikstra, P. S. Goldstein, and A. Webster
1995 Articial cranial deformation at the Omo M10 site:
A Tiwanaku complex from de Moquegua Valley,
Peru. Latin American Antiquity 6:145164.
Hrdlicka, A.
1912 Artificial deformation of the human skull with
especial reference to America. In Actas del XVII
Congreso Internacional de Americanistas, Sesin de
Buenos Aires, 1723 de mayo de 1910, edited by
Robert Lehmann-Nitsche (Secretario General del
Congreso), 147149. Buenos Aires: Imprenta de
Coni Hermanos.
Iscan, Y. M., and K. Kennedy, eds.
1989 Reconstruction of life from the skeleton. New York: Alan
R. Liss, Inc.
Jessup, D.
1987 Social and economic change in a post-imperial
society: The Chiribaya period of southern Peru.
Dissertation proposal, University of Chicago.
1990a Desarrollos generales en el Intermedio Tardo en el
Valle de Ilo, Peru. Unpublished Informe Interno del
Programa Contisuyu.
1990b Rescate arqueolgico en el Museo de Sitio de San
Gernimo, Ilo. In Trabajos arqueolgicos en Moquegua
Peru, edited by L. K. Watanabe, M. E. Moseley,
and F. Cabieses, 3:151165. Moquegua: Programa
L O Z A D A A N D B U I K S T R A 223
Contisuyo del Museo Peruano de Ciencias de la
Salud, Southern Peru Copper Corporation.
Johnson C., R. Gorlin, and V. Anderson
1965 Torus mandibularis: A genetic study. American Jour-
nal of Genetics 17:433442.
Larsen, C. S.
1997 Bioarchaeology: Interpreting behavior from the human
skeleton. Cambridge: Cambridge University Press.
Lozada, M. C., D. E. Blom, and J. E. Buikstra
1997 La deformacin craneana articial: Una prctica cul-
tural entre los pobladores Chiribaya de Ilo. Paper
presented at Seminario sobre Arqueologa de Ilo,
Ponticia Universidad Catlica del Per. Lima.
Lozada, M. C., R. Haydon, and, J. E. Buikstra
1993 Approaches to the study of genetic composition in
Late Intermediate period Chiribaya settlements,
southern Peru. Paper presented at the 59th Annual
Meeting of the Society for American Archaeology,
St. Louis, Missouri.
Manly, B.
1991 Randomization and Monte Carlo methods in biology.
London: Chapman and Hall.
Marcus, J.
1987 Prehistoric shermen in the kingdom of Huarco.
American Scientist 75:393401.
Netherly, P.
1977 Local level lords on the north coast of Peru. Ph.D.
dissertation, Department of Anthropology, Cornell
University.
Oakland Rodman, A.
1992 Textiles and ethnicity: Tiwanaku in San Pedro de
Atacama, north Chile. Latin American Antiquity 3:
316340.
Owen, B.
1993 A model of multiethnicity: State collapse, competition,
and social complexity from Tiwanaku to Chiribaya in
the Osmore Valley, Peru. Ph.D. dissertation, Depart-
ment of Anthropology, University of California, Los
Angeles.
Pietrusewsky, M., and M. Douglas
1993 The skeletal biology of an historic Hawaiian cem-
etery: Familial relationships. Homo 43:245262.
Rabinowitz, J.
1983 La lengua pescadora: The lost dialect of Chimu sh-
ermen. In Investigations of the Andean past., edited
by D. H. Sandweiss, 243267. Ithaca, NY: Cornell
Latin American Studies Program.
Ramrez, E. S.
1982 Retainers of the lords or merchants: A case of mis-
taken identity? In El hombre y su ambiente en los Andes
centrales, edited by L. Millones and H. Tomoeda,
123136.Senri Ethnological Studies 10. Osaka.
Richtsmeier, J. T., J. Cheverud, and J. Buikstra
1984 The relationship between cranial metric and non-
metric traits in the rhesus macaque from Cayo
Santiago. American Journal of Physical Anthropology
64:213222.
Rice, D. S.
1993 Late Intermediate period domestic architecture
and residential organization at La Yaral. In Domestic
architecture, ethnicity, and complementarity in the south-
central Andes, edited by M. S. Aldenderfer, 6682.
Iowa City: University of Iowa Press.
Rivera, M., and F. Rothhammer
1986 Evaluacin biolgica cultural de las poblaciones
Chinchorro. Nuevos elementos para la hiptesis de
contactos transaltiplnicos: Cuenca Amazonas-costa
Pacco. Chungar 16/17:295306.
Rostworowski de Diez Canseco, M.
1961 Curacas y sucesiones, costa norte. Lima: Imprenta
Minerva.
1970 Mercaderes del Valle de Chincha en la epoca prehis-
pnica: Un documento y unos comentarios. Revista
Espaola de Antropologa Americana 5:135177.
1975a Pescadores, artesanos, y mercaders costeos en el
Per prehispnico. Revista del Museo Nacional 41:
311349.
1975b Algunos comentarios hechos a las ordenanzas del
Doctor Cuenca. Historia y Cultura 9:119154.
1977 Etna y sociedad: Costa peruana prehispnica. Lima:
Instituto de Estudios Peruanos.
1981 Los pescadores del litoral peruano en el siglo XVI
Yunga Guaxme. Nova Americana 4:1142.
1989 Costa peruana prehispnica. Lima: Instituto de Estu-
dios Peruanos.
1990 Ethnohistorial considerations about the Chimor.
In The northern dynasties: Kingship and statecraft in
Chimor, edited by M. E. Moseley and A. Cordy-
Collins, 447460. Washington, D.C.: Dumbarton
Oaks Research Library and Collection.
1993 La voz parcialidad en su contexto en los siglos XVI y
XVII. In Ensayos de historia Andina, 231240. Lima:
Instituto de Estudios Peruanos.
1995 Historia del Tahuantinsuyu. Lima: Instituto de Estu-
dios Peruanos.
1997 Coastal islands of Peru: Myths and natural
resources. In Life and afterlife in ancient Peru: Trea-
sures from Rafale Larco Herrera Museum, 3339.
Catalogue of Objects. San Francisco: Fine Arts
Museums.
224 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
Rothhammer, F., J. A. Cocilovo, S. Quevedo, and E. Llop
1983 Anidad biolgica de las poblaciones prehistricas
del litoral Ariqueo con grupos poblacionales
costeros Peruanos y Altiplnicos. Chungar 11:
161166.
Rothhammer, F., S. Quevedo, A. Cocilovo, G. Focacci, and
E. Llop
1981 Microevolucin en poblaciones prehistricas del
rea Andina: Variacin craneomtrica cronolgica
en los Valles de Arica. Chungar 8:275289.
Rothhammer, F., S. Quevedo, J. A. Cocilovo, and E. Llop
1982 Microevolution in prehistoric Andean populations:
I. Chronologic non-metrical cranial variation in
northern Chile. American Journal of Physical Anthro-
pology 58:391396.
1984 Microevolution in prehistoric Andean populations:
II. Chronologic non-metrical cranial variation in
northern Chile. American Journal of Physical Anthro-
pology 65:157162.
1989 Orgenes y microevolucin de la poblacin
Chilena. In Culturas de Chile prehistoria: Desde sus
orgenes hasta los albores de la conquista, edited by
J. Hidalgo, V. Schiappacasse, H. Niemeyer, C.
Aldunate, and S. Ivn, 403413. Santiago de Chile:
Andrs Bello.
Rothhammer, F., and C. Silva
1989 Peopling of Andean South America. American Jour-
nal of Physical Anthropology 78:403410.
1990 Craniometrical variation among South American
prehistoric populations: Climatic, altitudinal,
chronological, and geographic contributions. Ameri-
can Journal of Physical Anthropology 82:917.
Rothhammer, F., C. Silva, J. A. Cocilovo, and S. Quevedo
1986 Una hiptesis provisional sobre el poblamiento de
Chile basada en el anlisis multivariado de medidas
craneomtricas. Chungar 16/17:115118.
Sallnow, M. J.
1991 Dual cosmology and ethnic division in an Andean
pilgrimage cult. In Pilgrimage in Latin America,
edited by N. Ross Crumrine and A. Morinis.
281306. New York: Greenwood Press.
Sandweiss, D. H., J. B. Richardon III, E. J. Reitz, Jeffrey T.
Hsu, and R. A. Feldman
1989 Early maritime adaptations in the Andes: Pre-
liminary studies at the Ring Site, Peru. In Ecology,
settlement, and history in the Osmore Drainage, Peru,
edited by D. S. Rice, C. Stanish, and P. R. Scarr,
1835. BAR International series 545(i). Oxford:
British Archaeological Reports.
Sanford, M. K., ed.
1993 Investigations of ancient human tissue. Langhorne:
Gordon and Breach Science Publishers.
Saunders, S., and F. Popovich
1978 A family study of two skeletal variants: Atlas bridging
and clinoid bridging. American Journal of Physical
Anthropology 49:193204.
Saunders, S. R.
1989 Non-metric skeletal variation. In Reconstruction of
life from the skeleton, edited by M. Y. Iscan and K.
Kennedy, 95108. New York: Alan R. Liss, Inc.
Shennan, S. J.
1989 Introduction. In Archaeological approaches to cultural
identity, edited by S. J. Shennan. 132. London:
Unwin Hyman.
Shimada, I.
1991 Pachacamac archaeology: Retrospect and prospect.
In Pachacamac: A reprint of the 1903 edition by Max
Uhle, edited by I. Shimada, xvilxvi. Philadelphia:
The University Museum of Archaeology and Anthro-
pology, University of Pennsylvania.
Soto, P., and F. Rothhammer
1975 Comparacin de pastrones de diversificacin
gentica y cultural entre poblaciones selvticas y
altiplnicas del Per. Chungar 5:8186.
Soto, P., F. Rothhammer, C. Valenzuela, E. Llop, and
Z. Harb
1975 Aplicacin de un mtodo de distancia gentica en
la comparacin de poblaciones prehispnicas de
Amrica. Chungar 5:7379.
Spence, M.
1974 Residential practices and the distribution of
skeletal traits in Teotihuacan, Mexico. Man 9:
262273.
Standen, V., and L. Nuez
1984 Indicadores antropolgicos-fsicos y culturales
del cementerio precermico Tiliviche-2 (norte de
Chile). Chungar 19:135155.
Stanish, C.
1992 Ancient Andean political economy. Austin: University
of Texas Press.
Susuki, M., and T. Sakai
1960 A familial study of Torus palatinus and Torus man-
dibularis. American Journal of Physical Anthropology
18:263272.
Sutter, R. C.
1997 Dental variation and biocultural afnities among
prehistoric populations from the coastal valleys
of Moquegua, Peru, and Azapa, Chile. Ph.D.
L O Z A D A A N D B U I K S T R A 225
dissertation, Department of Anthropology, Uni-
versity of Missouri, Columbia.
2000 Prehistoric genetic and culture change: A bioar-
chaeological search for pre-Inka Altiplano colonies
in the coastal valleys of Moquegua, Peru, and Azapa,
Chile. Latin American Antiquity 11 (1): 4370.
Tomczak, P.
1995 Paleodietary analysis of Chiribaya Alta: The stable
isotope and trace element evidence. Masters the-
sis, Department of Anthropology, University of
Chicago.
Torgersen, J.
1951 The developmental genetics and evolutionary
meaning of the metopic suture. American Journal of
Physical Anthropology 9:193210.
Wiessner, P.
1983 Style and social information in Kalahari San projec-
tile points. American Antiquity 49:253276.
226
I
n the early 1980s, a number of archaeologists
working in the south-central Andes focused on a
major methodological question in Andean economic
anthropology. That question surrounded the nature
of colonization in zonal complementarity models as
articulated by John V. Murra and several other schol-
ars in the 1960s, 1970s, and early 1980s. Murra had
convincingly demonstrated that late prehispanic and
early Colonial peoples in the central Andes had colo-
nized different ecological zones to complement their
economic base. These zones were dened largely by
altitude; hence the synonym verticality was applied
to these models to reect the stratication of ecologi-
cal regions in the Andes. In this model, these resource
zones were colonized by groups over great distances.
Unique to this model was the fact that the colonies
remained ethnically identical to each other. Unlike
colonies known from history and ethnography in other
parts of the world, the colonists coexisted with others
to form a highly complex political geography. Ethni-
cally similar groups of people were separated by huge
distances, and they lived next to people who were not of
their ethnic group who had also colonized the region.
The means by which resources moved over the
landscape in verticality models was therefore quite
novel in economic anthropology. Unlike contempo-
rary Western macroeconomic theory where individual
people exchange goods and labor as autonomous
agents, the economic relationships between colonists
in verticality models were dened by ethnicity. That
is, colonists were bound by traditional relationships
of kinship and community. Given the importance of
ethnicity in the zonal complementarity literature,
defining ethnicity in archaeological contexts is an
essential methodological issue. Migration, coloniza-
tion, and ethnicity have therefore been the focus of
much theoretical, methodological, and empirical
research in the Andes for decades.
The chapters in this volume represent the state-of-
the-art of the archaeology of zonal complementarity.
In this paper, I try to contextualize the contributions of
this paper in historical and theoretical context. From
CHAP T E R 1 3
Discussion:
MI GRAT I ON, COL ONI E S ,
AND E T HNI CI T Y I N T HE
S OUT H- CE NT RAL ANDE S
CHARLES S TANI S H
S T A N I S H 227
biological anthropological data to textiles, this collec-
tion of essays describes the means by which we can
address complex models of exchange in the archaeo-
logical record by dening ethnic afliations, migrations,
and colonization.
ZONAL COMPLEMENTARITY
In 1964, Waldemar Espinoza S. published a transcrip-
tion of the Visita of Garci Diez de San Miguel that was
conducted in 1567 (Garci Diez de San Miguel 1964
[1567]). Appended to the Visita was an article by Murra
(1964). In this article, Murra rst suggested his model of
verticality. In two critical publications in 1968 and 1972,
Murra restated the model of zonal complementarity,
establishing it as the principal theoretical framework for
analyzing Andean political economy for decades. The
basic principle behind this model is that the vertical
stratication of ecological zones in the Andes has pro-
foundly affected the political and economic strategies
of prehispanic populations. The traditional model is
characterized by the direct control of colonial lands by
polities outside of their core region or home territory
(Stanish 1992:3). Populations strategically locate their
colonies to control a diverse set of ecological zones.
This geographical pattern allows the various, comple-
mentary ecozones to be exploited by a single polity.
Hypothetically, the resulting distribution of colonies
creates an archipelago of landholdings over various
ecological zones. The overlap of archipelagos resulted
in a complex patchwork of different ethnic groups and
political units.
As early as the seventeenth century, Bernab Cobo,
perhaps the rst Andean scholar to recognize verticality,
described the relationship between the Andean environ-
ment and the political economy:
It is necessary to presuppose the existence of an ancient
custom of these people, and it is that when some prov-
ince did not produce certain foods, especially none of
their bread, which was maize, but was suitable for other
uses, special arrangements were made. For example,
due to the extreme cold, the provinces of Collao do
not produce maize or other seeds or fruits of temperate
lands, but they are very abundant in pasture lands and
most appropriate for raising livestock and producing
papas [potatoes], from which chuo, their substitute for
bread, is made, as well as some other roots. For the
inhabitants of these provinces, the Inca had picked out
lands which lie in the hot valleys of the seacoast on one
side and on the other side of the mountains toward the
Andes; in these temperate valleys they plant the crops
that they lack in their own lands . . .
(Cobo 1979 [1653]:192)
In zonal complementarity models, the home com-
munity would send out relatives to distant ecological
zones to exploit the ecological differences. The links
between the colony and home territory were very
strong. In fact, the principal cultural linkages between
distant territories consisted of nonmarket economic
exchange mediated through kinship. In this sense, the
archipelago distribution of colonies all tied together by
kinship resulted in a discontinuous distribution of the
same ethnic group. An example that Murra used was
the Lupaqa kingdom of the western Titicaca basin. The
Lupaqa were a sixteenth-century Aymara-speaking pol-
ity who had maintained a degree of autonomy in both
the Spanish and Inka states (Stanish 2000).
The Visita of Garci Diez de San Miguel of 1567
explicitly mentioned instances where the Lupaqa sent
out colonies to the western and eastern slopes to exploit
the lower altitude resources. A typical reference in the
Diez de San Miguel Visita is: each year the majority of
indians go to Sama, Moquegua, Capinota and Cuzco . . .
for maize, aji and other staples that do not grow in this
province and from which they barter cattle, cloth, wool
and charqui. (Garci Diez de San Miguel 1964 [1567]:
208). Valleys to the east such as Larecaja were included
with references to valleys in the west such as Sama, Lluta,
and Moquegua. The Visita even noted the number of
people that lived in the enclaves. The suggestion by both
the Lupaqa witnesses in the Visita as well as by Murra
was that zonal complementarity patterns were even more
prevalent during the Inka and pre-Inka periods.
The verticality model as presented by Murra is
consistent with the observation that central Andean
economies did not have any developed market mecha-
nisms. In Murras model, the relationship between
the home territories and their colonies was mediated
through mechanisms of redistribution and reciproc-
ity (Murra 1985:16). In the case of the Lupaqa, the
relationship between the elite of the home territory
and their colonies appears to have been dominated
by a redistributive relationship, characterized by the
Spanish historians as the payment of tribute. We also
can surmise that alongside this colonial relationship
were family-level reciprocal ones, with the lowland
colonies exchanging foodstuffs from the lowlands for
wool, freeze-dried meat (charqui), freeze-dried pota-
toes (chuo), lake sh, and other commodities from the
Titicaca region.
228 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
ETHNICITY AND ZONAL COMPLEMENTARITY
As described above, ethnicity is a central concept in
zonal complementarity models. This is because 1)
Murra viewed the home territory and the enclaves
as being ethnically identical, and 2) in a non-mar-
ket economy, exchange had to be mediated through
sociological mechanisms, particularly kinship. Unlike
anything documented on any scale in the ethnographic
record before, Andean peoples were purported to have
developed this unique adaptation to a unique envi-
ronment. Each home territory and colony was linked
through sociological ties. Markets did not exist prior
to the Spanish conquest, and therefore all trade was
through reciprocal and redistributive mechanisms as
outlined by the theoretical work of Karl Polanyi. The
work of Polanyi permeates the theoretical framework
of Murra and the zonal complementarity theorists
throughout the 1970s and 1980s. While Polanyi was
ambiguous as to whether redistribution, reciprocity,
market exchange, and so forth were simple mecha-
nisms or entire economic systems, Murra and his
colleagues utilized these theoretical tools well to
embed the Andean data into a viable, nonformalist
economic framework.
In the early 1980s, archaeologists started to seriously
test verticality models. I, along with others who began
work on the late prehispanic periods of the western
slopes of the south-central Andes, was confronted with
discriminating between colonial enclaves and sites that
engaged in economic exchange mediated by kinship.
In the colonial enclave case, the home territory and
the colony would be ethnically similar. In the case of
simple exchange relationships between different ethnic
groups located in different zones, we expected to see
some kind of ethnic differentiation between the sites
in the archaeological record.
The methodological problem facing us was how to
discriminate between different ethnic groups in the
archaeological record, as predicted by zonal comple-
mentarity models. As any anthropologist knows, and
as the papers in this volume indicate, dening ethnicity
even in living people is an immensely difcult task. Eth-
nicity is highly uid. Language, dress, food, and many
other customs dene ethnic afliations. At the group
level, intermarriage and in-migration confound any
attempt at creating tidy ethnic categories. Many of the
indicators of ethnicity do not survive in the archaeologi-
cal record. Finally, ethnicity changes over generations
as people adopt and reject some or most attributes of
their former ethnic group.
In spite of these difculties, we felt it absolutely
necessary to develop some methodological means to
test verticality models. We were fortunate to work in a
physical environment ideally suited to test such subtle
anthropological concepts. One of the great advantages
of the western slopes of the Pacic watershed is the
extreme aridity. Archaeological preservation is excel-
lent, and there is little natural soil formation. Since the
western valleys are deserts, village life away from the
coastline is only possible with some form of irrigation.
Since irrigated land is so precious, people throughout
prehistory and history tended to build their habitation
sites away from the bottomlands and on the dry slopes
above the rivers and canals. When the irrigation systems
collapsed, former habitations were often left stranded in
a virtually rainless, economically useless environment.
As a result, archaeological settlements were easily visible
on the surface. These settlements had intact structure
foundation walls, looted tombs, and substantial organic
remains on or near the surface.
HOUSEHOLD ARCHAEOLOGY AND THE
DEFINITION OF ETHNIC GROUPS
The ability to easily discern the walls of structures on
archaeological sites was a major factor in deciding to
use some of the concepts and techniques of household
archaeology to test models of zonal complementarity.
The dry environment of the western valleys preserves
virtually the entire architectural layouts of the sites.
Clusters of structures are visible, as are cemetery areas
and other functionally specic structures.
In one publication on the subject (Stanish 1992:34
41), I dened the household in methodological terms
as the basic coresidential economic unit in the settle-
ment. This term was intended to distinguish between
the household as a sociological unit represented by the
biological family and the household as an archaeological
unit represented by coresidential work units. In most
cases, the coresidents are biological family members,
but the converse is not true. It is likely that some bio-
logical family members live elsewhere on the settlement
or in other structures in neighboring areas.
I argued that the archaeological household can be used
to identify ethnic afliation in settlements. The theoreti-
cal underpinning of this proposition is that household
layout is largely culturally contingent and conservative.
Other aspects of the household construction and form
respond to local environmental conditions, such as
rainfall, temperature, altitude, and building resources.
S T A N I S H 229
However, the actual spatial distribution of functionally
distinct rooms and activity areas, as well as structure
shape, are largely culturally determined.
Numerous historical examples support this proposi-
tion, ranging from the Viking colony on Newfoundland
to English houses in Iquique on the coast of Peru.
Throughout the ethnographic and historical record,
there is good evidence demonstrating how migrant
populations will maintain the cultural principles mani-
fest in domestic architecture while using local materials
to build those structures and adapting to local circum-
stances where necessary. Adobe or sod may replace
wood, as in the case of European migrants to the
American West, but the spatial layout of houses tends
to reproduce the ethnic customs of migrants.
There are of course numerous counter examples of
migrant populations adopting new domestic architec-
ture. In particular, forced migrations into hostile areas,
or migrations to cities where the complex processes of
ethnogenesis occur, would not be appropriate analogs
for zonal complementarity models.
Zonal complementarity models, as articulated by
Murra and others, remain ambiguous about the nature
of colonies. I identied three potential colonial congu-
rations as suggested by the verticality literature (Stanish
1992:4345). The rst kind of colonization would be a
phenomenon in which an entire region, perhaps a valley
or section of a river valley, would be occupied by one
ethnic group. A second kind would be a case where eth-
nically distinct sites share a region with other ethnically
distinct sites. The third case would be one in which a
single settlement was multiethnic, with perhaps distinct
barrios in a large village or small town. The historical
documents were not clear on this, and Murra and his
colleagues were never explicit as to the precise nature
of the vertical settlement in their models.
Archaeological research in Moquegua demonstrated
that the rst two kinds of congurations were possible.
Whole regions were colonized by groups such as
the Tiwanaku in mid-Moquegua region (Goldstein
1993). Likewise, during the Otora period in an upper
Moquegua tributary of the same name, we found two
very distinct architectural types living side by side in
the same valley, ostensibly at the same time (Stanish
1989a, 1989b, 1992). However, we never discovered
a multiethnic settlement in the upper reaches of the
Moquegua area, although Richard Reycraft discusses a
possible multiethnic settlement at Carizzal. It is likely
that such multiethnic settlements are extremely rare
due to the complex processes of ethnic negotiation
that occur in such contexts. As anthropologists and
comparative sociologists have noted, it is precisely in
such cultural environments that new ethnic identities
emerge. This process of ethnogenesis is a hallmark of
urban environments and is decidedly not a characteristic
of zonal complementarity models.
This observation is reinforced by Garth Bawdens,
John Januseks, and Steve Bourgets contributions to this
volume. Bawden notes how the dominant ideology of
a political elite in an urban setting will be expressed in
markers of ethnicity. Bourget discusses elite iconography.
At Moche, much of the iconography was involved with
elite reinforcement of their status. In contrast, Janusek
identied distinctive areas at Tiwanaku where groups
maintained their earlier identities while adapting to a new
urban environment. In all three cases, the urban context
is profoundly different than that hypothesized in zonal
complementarity models at the village level.
DOMESTIC AND NONDOMESTIC CONTEXTS
AND THE DEFINITION OF ETHNICITY
Prior to our work in Moquegua, it was typical for
archaeologists to dene colonies based almost exclu-
sively on stylistic comparisons of artifacts, usually
pottery, between different sites in different areas. This
artifact-based approach was problematic for several
reasons. The most critical drawback, in my opinion,
was a reliance on stylistic comparisons between grave
goods. I argued that artifacts in tombs were not a good
context to dene the ethnic afliation of a settlement.
This was due to the fact that exotic goods of high value
tended to be buried with individuals in far greater
numbers than in other contexts. In support of this
proposition, I described results from Cigliano and his
colleagues (1973) and Pollard (1984) at the Late Inter-
mediate period site of Tastil in Argentina. In this case
almost 66% of the whole pottery vessels discovered in
the tombs were exotic, whereas only 15% of the whole
pottery found in the residential structures were exotic.
Even more dramatic results were obtained when all pot-
sherds were used. In this case, only 2% of all pottery
found in domestic contexts were exotic. Similar patterns
were found in my own excavations in the Otora Valley
(Stanish 1985), at the site of Playa Miller in Chile, and
in other sites where we had controlled excavations of
both habitation and cemetery areas.
In short, the only factor that explained this pattern-
ing was the context in which the artifacts were found.
Domestic contexts produced far fewer exotic goods than
230 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
did funerary contexts. Imagine a model of interregional
relationships and exchange developed from only one of
these contexts. Using only the funerary data, one would
argue for a multiethnic colony. Using only domestic
data, one would argue for an isolated settlement with
minimal interregional exchange.
I therefore proposed that as part of the methodol-
ogy to dene ethnically distinct colonies, we had to go
beyond household layout and control for the context
in which artifacts were found. In short, domestic
contexts (that included structure layout) tend to be
more conservative in preserving ethnic markers
than nondomestic contexts, particularly cemeteries or
tombs. This contextual approach was used to ana-
lyze settlements in the Otora Valley. Based upon this
methodology, I identied a uid cultural historical land-
scape beginning with an initial colonization by people
ethnically linked to the nal Tiwanaku occupation of
the region. Subsequent periods were characterized by
multiethnic colonization of the Otora Valley followed
by a process of ethnogenesis. By AD 1300 or so, the
ethnic afliation of the inhabitants had transformed
over time and resulted in a new group that was named
Estuquia by Michael Moseley.
Kevin Vaughns contribution in this volume has
expanded and improved on this methodological sug-
gestion in an admirable manner. He dened the round
structure, house-patio pattern in the Ica-Grande
region, while also recognizing a degree of variability
around this normative pattern. It is quite intriguing
that he is able to detect differences in settlement
patterns from valley to valley in the region. These
valleys are ecologically similar; as Vaughn points out,
the differences could reect some kind of ethnic dif-
ferentiation, but much more work must be conducted
before we can draw such conclusions.
In this light, the chapter by Richard Sutter chal-
lenges some of these assumptions that funerary
contexts are poor markers of ethnicity. His work
suggests, to the contrary, that pottery in tombs is
the best indicator in his work. This is a fascinating
pattern deserving of more research. Perhaps the fact
that his sample derives from a coastal population may
account for some of this variation. Certainly, his data
serve to support the use of ceramic style for dening
ethnic afliation and challenge some of my ideas on
this methodological approach.
In short, the archaeological evidence supports two
kinds of migration or colonization models in the late
prehispanic periods in the south-central Andes. As
dened above, these are either characterized by the con-
trol of a region or the control of individual sites within
a region. Multiethnic settlements were not discovered.
Over time, ethnicity shifted. The regional political and
economic relationships were uid as well. Acquiring
necessary ecologically specific goods was achieved
through a mixture of colonization and exchange.
LO ANDINO VERSUS COMPARATIVE
ANTHROPOLOGY
The model of zonal complementarity has been a
powerful theoretical tool used to understand Andean
societies. If one analyzes the historical context in which
it was created, one can see that Murra and others were
struggling to develop a viable Andean mode of produc-
tion that both conformed to general Marxist theory
and emphasized the uniqueness of Andean peoples
exploitation of their natural environment. The degree
to which verticality is different from other areas of
the world where resource zones shift due to altitude
remains problematic.
After two decades of research, I must reject this
central tenet of zonal complementarity theoryi.e.,
that it is a pattern and/or process unique to the Andes.
I argue instead that the cultural processes in the Andes
can be modeled like other areas of the world. In this
sense, I take a comparative approach and focus on the
processes of the development of political and economic
complexity and do not see zonal complementarity as
a uniquely Andean mode of production. Clearly, the
geographical characteristics of the central Andes identi-
ed by Murra profoundly affected the development of
complex society. However, by rejecting the uniqueness
of the verticality ideal (see Forman 1978; Van Buren
1996), we redirect our attention away from dening a
unique Andean mode of production to modeling the
anthropological processes that undergird the origins
and evolution of complex society in the central Andes.
I would also go so far as to suggest that the whole
concept of the Andes as a single cultural area is a
result of Inka and Spanish imperial policy (Stanish
2001). Prior to the Inka Empire, there were three
major linguistic, ecological, and cultural zones in the
central Andes: a central and north coastal zone where
Mochica and related languages were spoken, a north-
central and central Andean zone where Quechua and
related languages were spoken, and the south-central
Andes where Aymara and Puquina predominated. Each
of these areas had a relatively insular cultural history,
S T A N I S H 231
with Moche, Wari, and Tiwanaku emerging in the rst
millennium AD as pristine states. Obviously, there
is some borrowing of cultural elements between Wari
and Tiwanaku and Wari and Moche. But, this level of
borrowing appears to be no greater than that between
the Nile Valley and Mesopotamia in the late fourth
millennium BC. Few would argue that these two areas
were one cultural zone. Yet, Andean scholars combine
into one cultural area a region that, in the Old World,
would stretch from Cairo to the Indus Valley.
One could reasonably argue that the notion of a
transcendent Andean ethnicity that incorporated
many disparate languages and cultures was a product
of Inka and Spanish imperial needs. Such a process
is common in most empires around the world. The
need to create ideologies of power that unite disparate
regions of the world is common in all expansive state
systems. It is certainly true that the Inka state con-
sciously sought to make people mark their ethnicity
with appropriate dress. But, at the ofcial level of
state ideology, as detected in their theologies of origin,
one can see an attempt to unite their realm under a
common sense of unity. Of course, the Spanish Empire
had very similar needs, and this ideology of Andean-
ess, from Quito to Santiago de Chile, was preserved
in Spanish state strategies.
There are numerous features of the many Andean
cultures that are truly unique. In fact, most aspects of all
cultures are to a great degree historically contingent, such
as language, dress, feasting styles, kinship systems, and so
forth. But economic production and exchange, and the
kinds of political organizations that order the economy,
are subject to strict selective pressures. There is sub-
stantially less variation in political economy around the
world than there is in other aspects of culture. The use of
colonies and complex exchange relationships to acquire
access to nonlocal resources is common around the world
in both space and time. Zonal complementarity as an
economic strategy can be understood as one example of
the processes of political evolution seen throughout the
archaeological and ethnographic record.
ETHNICITY IN THE ARCHAEOLOGICAL RECORD
When we rst started our research in Moquegua in the
1980s, biological anthropology was just beginning to
develop the potentials of DNA analysis. I recall thinking
at the end of the decade, when I had nished my rst
book draft on zonal complementarity, that DNA analy-
sis would make most of the methodological advances
that we had achieved irrelevant in short order. In
fact, ancient DNA has not developed as quickly as we
would like. As Deborah Bloms ne paper demonstrates,
bioarchaeological approaches will not be replaced by
ancient DNA either. Her work with cranial deforma-
tion, a quintessential social modication of the human
body, shows that this classic marker of ethnicity in the
Andes can be effectively used to identify group identity.
As she also points out, ethnic identity is uid whereas
the deformation of the cranium at birth is permanent.
Cranial deformation data, combined with other kinds of
information, provides a very powerful means of dening
social complexities in the archaeological record.
The contributions of Amy Oakland Rodman,
Gioconda Fernandez Lopez, and Andrea Heckman
emphasize the role of textiles in marking ethnicity.
Unlike cranial deformation, textiles are a much more
uid form of ethnic identication. As a kind of social
skin, dress can be changed to suit the ethnic moment as
described by Oakland Rodman and Fernandez Lopez.
Heckman takes this logic a step further and sees modern
weavers actually place ethnically specic metaphors in
their products.
The denition of ethnicity in the archaeological
record still requires the intensive analysis of objects
and the context where those objects were found. The
papers in this volume nicely illustrate this observation.
Reycraft, for instance, suggests that textiles, domestic
architecture, tomb styles, and ceramic newares used
in ritual contexts provide good markers of ethnicity.
He tests these variables with data from Chiribaya
contexts and demonstrates how a multifactor approach
can provide unexpected, but exciting results. We still
use pottery styles in domestic contexts, iconography,
skeletal and other nongenetic biological information
(particularly cranial deformation), clothing, and tomb
design. Sloan Williamss paper illustrates the potentials
of ancient DNA, but also emphasizes how much basic
genetic work has to be completed before we can use it
effectively in assessing ethnicity in the archaeological
record. M. C. Lozada and Jane Buikstras paper elegantly
shows how non-metric cranial traits and cranial defor-
mation data can be used to sort out groups among the
archaeological populations of the Chiribaya. Biological
data will provide an additional source of independent
evidence to evaluate these models. After looking at the
rich data and meticulous interpretations in this volume,
I now realize that ancient DNA is not a panacea, but
one more tool in our archaeological kit that will help
us dene this complex phenomenon.
232 A R C H A E O L O G Y A N D E T H N I C I T Y I N T H E A N D E S
This volume brings together the whole range of
approaches that we can use to dene ethnic differ-
ences in the archeological record. Of course, the very
denition of ethnicity will change as we develop new
theoretical concepts. However, the work presented
in this volume stands as a model for integrated,
multidimensional methodologies in contemporary
anthropological archaeology.
REFERENCES
Cigliano, E. M.
1973 Tastil: Una ciudad preincaica Argentina. Buenos Aires:
Ediciones Cabargon.
Cobo, B.
1979 [1653] History of the Inca Empire: An account of the Indi-
ans customs and their origin together with a treatise on
Inca legends, history, and social institutions. Translated
and edited by R. Hamilton. Austin: University of
Texas Press.
Forman, S.
1978 The future value of the verticality concept: Impli-
cations and possible applications in the Andes. Actes
du XLIIe congrs international des amricanistes, 4:
233256. Paris.
Garci Diez de San Miguel
1964 [1567] Visita hecha a la provincia de Chucuito por Garci
Diez de San Miguel en el ao 1567. Lima: Ediciones
de la Casa de la Cultura del Per.
Goldstein, P.
1993 Tiwanaku temples and state expansion: A Tiwan-
aku sunken-court temple in Moquegua, Peru. Latin
American Antiquity 4 (1): 2247.
Murra, J. V.
1964 Una apreciacin etnolgica de la Visita. In Visita
hecha a la provincia de Chucuito por Garci Diez de San
Miguel en el ao 1567. Lima: Ediciones de la Casa
de la Cultura del Per.
1968 An Aymara kingdom in 1567. Ethnohistory 15:
115151.
1972 El control vertical de un mximo de pisos ecolgi-
cos en la economa de las sociedades Andinas. In
Visita de la provincia de Len de Hunuco en 1562,
427476. Documentos por la Historia y Etnologa
de Hunuco y la Selva Central 2. Hunuco: Univer-
sidad Nacional Hermilio V.
1985 The limits and limitations of the vertical archipel-
ago in the Andes. In Andean ecology and civilization,
edited by S. Masuda, I. Shimada, and C. Morris,
1520. Tokyo: University of Tokyo Press.
Pollard, G. C.
1984 Interregional relations in the southern Andes: Evi-
dence and expectations for understanding the late
prehistory of northwest Argentina and north Chile.
In Social and economic organization in the prehispanic
Andes, edited by D. L. Browman, R. L. Burger, and
M. A. Rivera, 205247. BAR International Series
194. Oxford: British Archaeological Reports.
Stanish, C.
1985 Post-Tiwanaku regional economies in the Otora
Valley, southern Peru. Ph.D. dissertation, Depart-
ment of Anthropology, University of Chicago.
1989a An archaeological evaluation of an ethnohistorical
model. In Ecology, settlement, and history in the Osmore
Drainage, Peru, edited by D. Rice, C. Stanish, and
P. R. Scarr, 303320. BAR International Series 545.
Oxford: British Archaeological Reports.
1989b Household archaeology: Testing models of zonal
complementarity in the south-central Andes. Ameri-
can Anthropologist 91 (1): 724.
1992 Ancient Andean political economy. Austin: University
of Texas Press.
2000 Negotiating rank in an imperial state: Lake Titi-
caca basin elite under Inca and Spanish control. In
Hierarchies in action: Cui bono? edited by M. Diehl,
317339. Occasional Paper 27. Carbondale, IL:
Southern Illinois University Center for Archaeo-
logical Investigations.
2001 The origins of the state in South America. Annual
Reviews in Anthropology 30:4164.
Van Buren, M.
1996 Rethinking the vertical archipelago, ethnicity,
exchange, and history in the south-central Andes.
American Anthropologist 98 (2): 338351.
233
INTRODUCTION
As the papers in this volume attest, the Andean past,
with its complex histories and rich archaeological
record, is an ideal context for the study of ethnicity. Yet,
the archaeological study of ethnicity is itself a complex
and contested arena. The papers presented here reect
numerous denitions of ethnicity, ranging from Kevin
Vaughns social/psychological construct, through Richard
Sutters contextual emphasis, to Garth Bawdens explicit
linkage of self-identication with social reproduction.
Some authors eschew explicit foci upon ethnicity, pre-
ferring to distinguish us and them in terms of social
identity (Janusek) or to include ethnicity within a more
broadly based consideration of biocultural groups (Sut-
ter). Emphasis upon context is pervasive, with some
authors including ethnohistoric models (Lozada and
Buikstra, Blom), while others focus upon iconography
(Bourget) and the potential that textiles hold for the
study of ethnicity in Andean contexts (Oakland Rod-
man and Fernandez Lopez, Heckman). The material
symbols chosen to represent ethnicity range from those
common to residential contexts (Vaughn) to combina-
tions of household and public monuments (Bawden).
And as a cautionary tale, Sloan Williams emphasizes the
importance of evaluating ancient kinship models against
empirical data derived from small-scale societies.
As described in the following section, recent archaeo-
logical approaches to ethnicity are commonly grounded
in a variety of social theories that permit us to be
distinguished from them. The lack of theoretical and
methodological consensus among the authors of these
chapters thus parallels the multiplicity of approaches
common to the eld today. This volume is, however,
somewhat unusual in its attempt to meld inherited and
culturally dened physical attributes, such as cranial
deformation, with variation in material culture. Just as
these authors argue for grounding ethnicity in histori-
cal traditions, the following section will briey sketch
historical developments in social and archaeological
theories that help contextualize their current contri-
butions. This is followed by an overview of key themes
developed in individual chapters.
CHAP T E R 1 4
Discussion:
E T HNOGE NE S I S AND E T HNI CI T Y
I N T HE ANDE S